The author(s) shown below used Federal funds provided by the U.S. Department of Justice and prepared the following final report: Document Title: Presumptive Field Testing Using Portable Raman Spectroscopy Author(s): Stephana Fedchak Document No.: 244564 Date Received: January 2014 Award Number: 2010-DN-BX-K201 This report has not been published by the U.S. Department of Justice. To provide better customer service, NCJRS has made this Federallyfunded grant report available electronically. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy Research and Development on Instrumental Analysis for Forensic Science Award Number 2010-DN-BX-K201 Final Technical Report Author: Stephana Fedchak Las Vegas Metropolitan Police Department This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Abstract The Las Vegas Metropolitan Police Department (LVMPD) currently utilizes commercially prepared chemical color test kits that officers use to presumptively identify cocaine, methamphetamine, and marijuana in the field. Over the past few years, false positive results have been discovered due to subjectivity of color interpretation and tedious procedures. Recognizing the need to find a more reliable method for presumptive field testing, the LVMPD Forensic Laboratory began investigating the use of Raman spectroscopy. The laboratory used National Institute of Justice funds to research and enhance existing Raman field technology to provide presumptive analysis of controlled substances. Part of the research focused on evaluating the performance of the ReporteR device, manufactured by SciAps, Inc. (previously DeltaNu, Inc.). Research was successful for analyzing methamphetamine and cocaine through the implementation of improvements designed to enhance the ReporteR device. The laboratory also investigated the advantages of algorithm-based fluorescent baseline correction technology by integrating a Raman microscope into the research design. To examine the specificity of the handheld device and validate its accuracy, the laboratory also used the microscope to compare results generated by the ReporteR. Casework evidence samples were analyzed in three distinct phases, comparing the ReporteR to the current chemical-based field test kits. The first phase of testing consisted of solely laboratory-based evaluation of the ReporteR. In the next two phases, narcotics detectives were recruited to perform tests on substances encountered in the field. From the onset of testing until after the enhanced ReporteR was evaluated, methamphetamine field testing accuracy increased 19.4%. Cocaine field testing accuracy increased 26.6% after enhancements. The reliability of the ReporteR was investigated through random re-testing of samples and results were consistent in approximately 87% of instances. Correlation settings and library content were studied and it was found that proper regulation of these factors can decrease erroneous results. Alleviating fluorescence issues of commonly encountered drugs is important to the future of Raman-based field testing. Unlike other techniques, the advantage of algorithm-based fluorescence baseline correction technology is that it could be translated to a portable system. During the third phase, the laboratory used the Raman microscope to determine if a specific patented algorithm would aid in the identification of fluorescing substances (e.g. ecstasy tablets, black tar heroin, and marijuana). Repeatable spectra indicating the presence of cannabinoids in Page i This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 marijuana plant material were collected using the Bruker Senterra Raman microscope. Testing ecstasy evidence revealed that substances can be identified, regardless of color and binding agents. A few synthetic drug analog reference materials were also evaluated. The compounds did produce spectra, however, in some instances; fluorescence caused hindrances in which the baseline correction technology was essential. Typically, testing black tar heroin directly is not feasible with Raman technology. In an attempt to remedy this, one technique involved the use of graphene to quench the fluorescence. The ReporteR device revealed possible indications of heroin peaks with the addition of graphene. An evaluation of cocaine/diluent mixtures using the Raman microscope demonstrated that baseline correction is valuable to the differentiation of mixture components. By testing the handheld devices in the lab, in the field, and by utilizing a Raman microscope, the LVMPD Forensic Laboratory has determined that portable Raman technology is an effective and reliable tool to presumptively identify methamphetamine and cocaine in the field. The laboratory has also identified which characteristics of portable Raman technology need enhancement in order to be implemented in a presumptive field testing program. Enhancing and implementing a new method of presumptive field testing will benefit law enforcement, forensic laboratories, and the court system by increasing safety, decreasing costs and time, decreasing the incidence of false positive results, and expediting the judiciary process. Page ii This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table of Contents Executive Summary……………………………….. 1 I. Introduction……………………………………… 9 1. Statement of the Problem…………………….. 9 2. Literature Citations and Review……………… 11 3. Rationale for the Research……………………. 16 II. Methods………………………………………… 18 III. Results…………………………………………. 32 IV. Conclusions……………………………………. 78 1. Discussion of Findings………………………... 78 2. Implications for Policy and Practice………….. 85 3. Implications for Further Research…………….. 87 V. References……………………………………… 89 VI. Dissemination of Research Findings………….. 95 Page iii This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Executive Summary In order to develop a new field testing program and enhance current Raman-based field technology, casework evidence samples were analyzed in three distinct phases, comparing the ReporteR to the current chemical-based field test kits. The first phase of testing consisted of solely laboratory-based evaluation of the ReporteR on suspected methamphetamine and cocaine samples. In the next two phases, narcotics detectives were recruited to perform tests on substances encountered in the field. During the third phase, the laboratory used a Bruker Senterra Raman microscope to determine if the patented convex function ‘f’ (also known as concave rubber band correction) algorithm would aid in the identification of fluorescing substances (e.g. ecstasy tablets, black tar heroin, and marijuana) and clarify spectra exhibiting a high background. A few synthetic drug analog reference materials were also evaluated using the Raman microscope. Enhancements were made to ReporteR based on the testing performed and were evaluated in the field and laboratory. • Before enhancements, methamphetamine field testing accuracy was 77.9%. • After enhancements, methamphetamine field testing accuracy was 100%. • Before enhancements, cocaine field testing accuracy was 73.4%. • After enhancements, cocaine field testing accuracy was 100%. • The reliability of the ReporteR was investigated through random re-testing of samples and results were consistent in approximately 87% of instances. • Correlation settings and library content were studied and it was found that proper regulation of these factors can decrease erroneous results. • Repeatable spectra indicating the presence of cannabinoids in marijuana plant material were collected using the Raman microscope and patented baseline correction. • Testing ecstasy evidence revealed that substances can be identified, regardless of color and binding agents using the Raman microscope and patented baseline correction. • Spectra was collected for synthetic drug analogs, however, in some samples where high background caused hindrances, the baseline correction technology was essential. • An evaluation of cocaine/diluent mixtures using the Raman microscope demonstrated that fluorescence baseline correction is valuable to the differentiation of mixture components. Page 1 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 • Testing black tar heroin directly is not feasible with current Raman technology, regardless of baseline correction. Enhancing and implementing a new method of presumptive field testing will benefit police officers, law enforcement field workers, forensic laboratories, and the court system. These benefits include safety, cost saving, time saving, and expedition of the judiciary process. A new method of field testing will enhance the ability of the law enforcement community to identify, analyze and interpret controlled substance evidence using a robust tool that gives immediate and defendable preliminary results for frequently encountered controlled substances. This scientific evaluation and justification for the court system is of utmost importance. The handheld device can give an officer the results he needs to present in court, but the court must first understand that it is a valid and reliable tool in preliminary identification of controlled substances. This will be accomplished by presenting extensive research to support such a proposal. Preliminary field testing was implemented by the Las Vegas Metropolitan Police Department in order to combat the rapidly increasing caseloads due to the proliferation of drug use across the population. This program provides the police department with a method to presumptively identify controlled substances in the field and present those findings in court, in lieu of laboratory analysis. The presumptive field testing program has been beneficial for many years, but the discovery of noncontrolled substances yielding false positive chemical field test results for methamphetamine and cocaine prompted the LVMPD Forensic Laboratory to research a superior field testing system for police officers to preliminarily identify controlled substances. With over 34,000 items of evidence being field tested each year, the elimination of the field testing program would overwhelm the laboratory and compromise due process. In 2009, the LVMPD spent $42,293 for color test field kits in order to presumptively identify methamphetamine, cocaine, and marijuana. The yearly expenditure of maintaining a Raman handheld unit is far less than the purchase of the color test field kits, with the one-time cost of a handheld unit being approximately $20,000. Besides subjectivity, the current chemical field testing method has other important drawbacks that further elevate the need for a new field testing method. The premade kits contain hazardous chemicals and require the police officers to directly handle the evidence and measure required testing amounts, which has possible safety and contamination considerations. Page 2 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Since 2008, the LVMPD Forensic Laboratory has been working with SciAps, Inc. (previously DeltaNu, Inc.), the manufacturer of a handheld Raman spectrometer, to devise a new preliminary field test for methamphetamine. The forensic laboratory’s research has shown the portable Raman method has several benefits over the existing field test. These include increased specificity, objectivity, non-destructive sampling, elimination of direct handling of drugs and hazardous chemicals, decreased human error, digital recording of results, and decreased cost to the department on a long term basis. The obvious shortcoming to using any type of Raman spectrometer is the interference of fluorescence. Fluorescence encountered during Raman analysis is inherent to many substances and many of these compounds are routinely submitted to the LVMPD laboratory. It was of particular interest to assess whether algorithm-based fluorescence baseline correction technology would assist in the analysis of such drug evidence. It is understood that current portable Raman technology may not be able to directly measure marijuana, ecstasy, or black tar heroin, as powerful fluorescence rejection tools are not normally applied or available with portable testing systems. A Raman microscope with an automated or semi-automated fluorescence correction option was essential in this research process to determine if this specific fluorescence baseline correction technology will aid in the identification of cocaine mixtures, black tar heroin, ecstasy, and marijuana. For methamphetamine and cocaine studies, a parallel study of the spectra collected by the handheld device versus the Raman microscope provided the data necessary to identify specific spectral variances that are unique to the substances being tested. This is intended to encourage manufacturers of portable Raman technology to further enhance devices that include immediate automatic baseline correction. Six portable ReporteR devices (manufactured by DeltaNu, Inc. / SciAps, Inc.), chemical color test field kits (manufactured by ODV), and a Bruker Senterra Raman microscope were purchased for use throughout the research. During the studies, LVMPD lab-created libraries were loaded onto the handheld devices and tested by both the field officers and in the laboratory. To decrease the incidence of erroneous results, test the specificity and selectivity of the handheld unit, and to simplify the testing procedure, each sample was tested no more than three times with a ReporteR through 2 mil polyethylene (the most commonly seen) plastic bags. In the first phase of methamphetamine and cocaine testing, adjudicated and current casework evidence samples were analyzed solely by the laboratory with the ReporteR and Page 3 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 chemical field test kits. In the second phase of methamphetamine and cocaine testing, prior to receiving the microscope, analysis was completed on recently seized case samples tested in the field and in the laboratory using the ReporteR, and chemical field test kits. Narcotics detectives from varied assignments (LVMPD Narcotics, an outside jurisdiction, an interdiction team, and a federal task force) were trained on the use of the ReporteR and testing procedures. The detectives performed tests with the ReporteRs on white to off-white, powdery, and chunky substances encountered in the course of duty. The third phase of methamphetamine and cocaine testing included the incorporation of the Raman microscope. Each evidence sample tested during this period was subjected to field testing with chemical field test kits and a ReporteR, and laboratory testing with a ReporteR, chemical field test kits (cocaine only), and the Raman microscope. All samples tested throughout the research project were confirmed via analysis with gas chromatography/mass spectrometry (GC/MS). A side by side comparison of selected methamphetamine and cocaine spectra collected during the research was performed using the OPUS software (Bruker, Inc.). In the first phase of laboratory methamphetamine testing, the ReporteR accuracy rate was 92%. Chemical color field testing of suspected methamphetamine evidence in the lab and in the field resulted in an accuracy rate of 100%. It was determined, after receiving these results, and finding that the prevalence of isopropylbenzylamine had waned, that methamphetamine chemical field test kits would no longer be performed in the remaining laboratory portion of the testing (but would continue in the field) unless a false positive field result was discovered. The second phase of methamphetamine ReporteR field testing revealed an accuracy rate of 77.9%. The lab accuracy rate was 100%. In the third phase of testing, suspected methamphetamine cases were tested in the field using an enhanced ReporteR with an accuracy rate of 100%. The laboratory accuracy rate for testing the same samples with a non-enhanced ReporteR was 96% during this phase. In the first phase of laboratory cocaine testing, the ReporteR accuracy rate was 73%. Chemical color field testing of suspected cocaine evidence in the lab and in the field also resulted in an accuracy rate of 73%. Due to these results, it was determined that cocaine chemical field tests would continue to be performed in the remaining laboratory portion of the testing. In the second phase of testing, ReporteR results for items suspected to contain cocaine had an accuracy rate of 73.4% for both officer and lab testing. In the third phase of testing, a total of 27 suspected Page 4 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 cocaine cases were tested in the field using an enhanced ReporteR with an accuracy rate of 100%. The laboratory accuracy rate for testing the same samples with a non-enhanced ReporteR was 66% during this phase. After the first and second phases of methamphetamine and cocaine testing, the LVMPD Forensic Laboratory requested that the manufacturer of the ReporteR (DeltaNu, Inc. / SciAps, Inc.) make specific software and firmware changes to the device to further enhance its usability. The final enhancements included a small portable Bluetooth printer that produces a customizable printout, enhanced software that has an improved signal-to-noise ratio, a newly designed polystyrene standard that prevents an officer from having to switch attachments in order to perform a self-test, a forced self-test in which an administrator can set the length of time allowed between self-tests, a new results display incorporating a number correlation rather than a bar graph, and a stronger magnet that secures the removable sampling attachment to the device. After these enhancements were applied to a ReporteR device and implemented in the field, they were well received by the detectives using the equipment. The enhancements also made the presumptive field testing system more efficient by eliminating the need for forensic scientists to perform scheduled calibration checks. Recording the results was also made more manageable by way of a physical print out. As is evident from both methamphetamine and cocaine ReporteRbased field testing result trends during the third phase of testing, the enhancements also appeared to increase testing accuracy. To test the reliability of the ReporteR devices, confirmed cocaine and methamphetamine samples previously tested during the first phase of laboratory testing were randomly selected and retested using the ReporteRs. The combined results for methamphetamine and cocaine were consistent in approximately 87% of instances. Although one false positive (benzoylecgonine) for cocaine had been discovered during laboratory reference material testing, the occurrence of this substance in evidence submissions is rare and encountered under unique circumstances. The benzoylecgonine false positive was discovered during ReporteR result correlation testing. These correlations translate to a bar graph in which intensity determines whether a match or “No match” result is displayed. By varying the level of an acceptable correlation, an administrator can somewhat regulate the number of false negatives, or false positives, that are received. A few false positive mismatches occurred during the third phase of methamphetamine testing (methamphetamine samples resulting in low Page 5 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 correlation cocaine matches), but these were determined to be due to equipment malfunctions and did not reoccur after the equipment was repaired. Preventative measures to deter these types of results from being submitted to preliminary court hearings are two-fold. (1) The correlation level could be set such that a result under a certain correlation could not be used for court purposes, and (2) Require that suspected cocaine samples that were discovered under wet or degraded conditions be submitted directly for laboratory analysis. An evaluation of cocaine versus commonly accompanying diluents was completed using the Raman microscope. Various spectral interferences were discovered and can explain the notable incidence of false negative results when testing cocaine mixtures with the ReporteR. It was evident that baseline correction capabilities are valuable to the study of the cocaine and diluent mixtures. Analysis of plant materials using Raman instrumentation has been regarded as difficult or impossible by researchers in the past, however, with recent advances, plant material may be analyzed non-destructively using this technology. Marijuana plant material casework samples from ten different events and two suspected marijuana “wax” samples were tested with the Senterra Raman microscope at 785 nm on aluminum foil slides. Secondary negative plant material samples were also examined using the Raman microscope. No viable spectra (spectra with peaks or features that could be differentiated) were obtained during plant material and tetrahydrocannabinol testing with the ReporteR. Several attempts were made on many samples to determine if there were actual microscopic features of the plant material that yielded viable spectra. Such features were identified. These features are best described as globular, amber colored resinous structures that can be smooth or striated in appearance. Repeatable spectra from marijuana plant material were collected using the Bruker Senterra Raman microscope and spectral attributes in common with cannabinoids were evident, specifically cannabinol and delta-9-tetrahydrocannabinol. Baseline correction was indispensable when detecting cannabinoids in marijuana. Although the probability of being able to use non-microscope based Raman technology on marijuana samples in the field is unlikely, the idea of detecting tetrahydrocannabinol and cannabinol in situ using spectroscopy opens the door for another confirmatory forensic drug testing method. As with marijuana, testing black tar heroin directly is not feasible with Raman-based portable technology. Various methods were investigated using both the microscope and the ReporteR. One technique included the use of in-house created graphene based substrates to quench the Page 6 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 fluorescence of the dark, black-brown substance. The Raman microscope did not produce any viable spectra during the graphene/heroin testing. The graphene testing did not produce any welldefined, detailed spectra with the handheld device, however, possible low intensity peaks in common with a heroin reference material were indicated, but this could not be confirmed. Ecstasy tablets are also routinely encountered by the LVMPD Forensic Laboratory. Contrary to marijuana and black tar heroin, testing ecstasy tablets directly is possible with Raman-based portable technology, but limited. Ecstasy tablets pose additional problems with Raman testing as they are often mixtures of several different compounds, come in a variety of colors, and could contain multiple binding agents. These may not only cause fluorescence issues, but mixture interferences as well. Limited testing on ecstasy tablets was completed. The testing of such items revealed that spectra of the diluents present in the tablets are more likely to be collected, however if focused on a crystal within the tablet that contains methylenedioxymethamphetamine, it can be detected by the microscope, even on colored tablets. ReporteR testing on selected samples did not collect viable spectra, only a high baseline. Various synthetic drugs were also tested as the proliferation of these compounds has caused recent concerns. A few designer synthetic drug analog reference materials were evaluated. Viable spectra were obtained; however, it was apparent that the amount of fluorescence or other background interference was dependent on isomeric form. Although it is unlikely that a field testing program will be useful in identifying the emerging synthetic drug analogs (i.e. cathinones, phenethylamines, cannabimimetics) due to their structural attributes and propagation, indications that these drugs may be present can be determined by portable Raman spectroscopy. As the manufacture of new synthetic drugs continues to increase, it is imperative that a method of assisting law enforcement in detecting such substances is available. As these analogous substances come and go from the illicit drug market on a daily basis, the identification of them is daunting. If these drugs can, in the very least, be detected by their base structure in the field using portable Raman spectroscopy, it would aid officers in their duties, even if the results could not be used in court. As the standard chemical color tests cannot differentiate between such compounds, and Raman technology can differentiate isomers, exploring the use of a portable instrumental based testing system is worth the time and consideration. The use of the baseline correction technology incorporated with the Bruker microscope system was essential in the collection of spectra of compounds that inherently produce fluorescence, namely Page 7 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 marijuana, ecstasy, and synthetic designer drugs. A high baseline was also evident in some methamphetamine and cocaine spectra when tested using the ReporteR, and baseline correction technology was useful in increasing the library matching capabilities of the OPUS software. Automatic baseline correction may not prove valuable against cocaine/caffeine mixtures, but it may help in deconvoluting spectra containing other contaminants. An algorithm based method would also be useful to assist in presumptive Raman analysis of components in cocaine mixtures. The development and implementation a successful field testing program to replace or, at the very least, reduce the use of chemical field test kits was the ultimate goal of this research. To determine if testing typically difficult samples with the fluorescence correction algorithm is achievable, and to explore the possibilities of applying this technology to a handheld device was also an important portion of the research. Improving handheld Raman technology and exploring the possibilities of alleviating the fluorescence issues of commonly encountered drugs is important to the future of field testing using Raman technology. With the research completed thus far, we have determined what aspects are important to implementing an instrumental based field testing program and which characteristics of portable Raman technology need enhancement in order to increase its user friendly attributes, efficiency, selectivity, and reliability. By testing the handheld devices in the lab, and in the field by narcotics officers, we have collected data in order to implement a reliable field testing method for methamphetamine and cocaine using Raman technology. Page 8 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 I. Introduction 1. Statement of the problem Preliminary field testing was implemented by the Las Vegas Metropolitan Police Department in order to combat the rapidly increasing caseloads due to a sudden increase in population, bringing along with it a sudden increase in drug use. The field testing program provided the LVMPD with multiple benefits. The laboratory did not have to employ extra scientists in order to handle the increased caseload. The program accelerated the judiciary process by accepting field test results at the preliminary hearing level. As the police officers are required to weigh the evidence prior to preliminary testing, the program assists them in determining which charges to file when booking the evidence. The presumptive field testing program has been beneficial for many years, but the discovery of a non-controlled substance yielding a positive color test result for methamphetamine in submitted evidence at LVMPD and other laboratories [2, 5] prompted the LVMPD Forensic Laboratory to find a better field test for police officers to preliminarily identify methamphetamine. Laboratory tests have also confirmed that presumptive tests performed in the field revealed false positive results for street samples purported as, but not containing, cocaine. With this information, it is possible that the Clark County, Nevada District Attorney’s Office will determine that the current field testing methods are no longer sufficient, in lieu of conclusive laboratory analysis, for preliminary hearings. With over 34,000 items of evidence being field tested each year, the elimination of the field testing program would overwhelm the laboratory and compromise due process. Law enforcement agencies spend tens of thousands of dollars on chemical-based color test field kits every year. In 2009, LVMPD spent $42,293 for color test field kits. The yearly expenditure of maintaining a Raman handheld unit is far less than the purchase of the color test field kits, with the one-time cost of a handheld unit being approximately $20,000. The handheld device may need replacement of its laser after 3 to 5 years, but at an estimated $2,500, it is a fraction of the yearly cost of color test field kits. This is a major consideration for most policing agencies, especially in the current economic climate. In addition, the training for this technology could be cost effective and easily obtainable by producing online video instruction. The current chemical field testing method has other important drawbacks that further elevate the need for a new field testing method. The premade kits contain hazardous chemicals and require the police officers to directly handle the evidence and measure out required testing amounts. This Page 9 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 makes for a dangerous, messy, and destructive presumptive test. A Raman device has far fewer safety considerations, can test through plastic packaging, and only takes the press of a button to get immediate results. Since 2008, the Forensic Laboratory has been working with SciAps, Inc. (previously DeltaNu, Inc.), the manufacturer of a handheld Raman spectrometer, to devise a new preliminary field test for methamphetamine and cocaine. After some trial and error, and with enhancements to the device, the forensic laboratory’s research has shown the Raman method to have several benefits over the existing field test. These include increased specificity, objectivity, non-destructive sampling, elimination of direct handling of drugs and hazardous chemicals, decreased human error, digital recording of results, and decreased cost to the department on a long term basis. Although shortcomings will have to be properly examined, the advantages of portable Raman technology over the current chemical preliminary field testing outweigh the disadvantages. The LVMPD realized that not only would certain enhancements to the portable device for use in the field be necessary, but an actual field testing system would need to be developed. Currently, using the chemical field test kits, an officer must complete a checklist and locate another field test certified officer to witness the test and sign the form as proof that testing was completed. This type of system creates room for human error. Certain enhancements such as adding a portable printer to the device so that an officer can include traceable results with the evidence at the time of booking, is one way to improve functionality and efficiency of a field testing program. The obvious limitation when using any type of Raman spectrometer is the interference of fluorescence. Fluorescence is inherent in many controlled substances [4, 16, 36]. In Nevada, ecstasy, cocaine, and heroin are frequently encountered controlled substances and are routinely submitted to the LVMPD laboratory for analysis. The predominant type of heroin encountered by Nevada law enforcement is in the form known as “black tar” and is dark black-brown in color. Ecstasy tablets come in all shapes, sizes, and colors. Cocaine is often diluted with substances that cause various spectral interferences [3, 25, 26, 35]. Each of these drug forms can be difficult to analyze using Raman technology due to the fluorescent background and high baselines they can produce. Current and future advents to current technology may help to solve these issues. Analysis of plant materials using portable Raman technology has previously been regarded as impossible. Fluorescence is an obvious culprit as to why testing would be difficult, but as so many chemicals are present in plant material, a microscope is, at this time, required for analysis to identify Page 10 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 certain areas of a plant that would contain the compound of interest. As marijuana is included in the LVMPD’s field testing program, the ability to test it in the field using instrumental based technology rather than field color test kits would be advantageous. With recent advances, researchers have shown that it is possible to analyze plant material using laboratory grade instrumentation [18-24]; however, the collection of spectral data of cannabinoids from marijuana plants has not been thoroughly studied. Improving handheld Raman technology and exploring the possibilities of alleviating the fluorescence issues of commonly encountered drugs is important to the future of field testing using Raman technology. A field testing program that will replace chemical-based testing methods would not only improve the preliminary testing system and judicial system for Clark County, Nevada, but may also serve as a model benefiting other jurisdictions with budgetary, staffing, and backlog issues. 2. Literature Citations and Review Color tests are useful tools in identifying the presence of a particular molecular moiety in groups of compounds [1]. Opiates like heroin and oxycodone, stimulants like methamphetamine and amphetamine, and hallucinogens like psilocin and lysergic acid diethylamide (LSD) represent groups of related compounds, or those that yield similar color test results [12]. In the early stages of controlled substance analysis, positive color test results were considered definitive and conclusive identification for controlled substances [1]. Color testing, now considered a preliminary “Category C” technique by the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG), is the principle method of law enforcement presumptive field testing of controlled substances. With the explosive, yet fleeting, emergence of the non-controlled substance isopropylbenzylamine into the street market, the non-specific nature of color tests became immediately more pertinent to the LVMPD Forensic Laboratory and the Drug Enforcement Administration (DEA) [2, 5]. The street form of isopropylbenzylamine was a crystalline substance visually similar to crystal methamphetamine. The Marquis field color tests (in this instance the Marquis test indicates the presence of amines) [12] yielded color changes similar to those obtained with methamphetamine and amphetamine [2]. The sodium nitroprusside color test is the second step in the methamphetamine field test, and differentiates between primary and secondary amines (e.g. amphetamine and methamphetamine) [12]. When isopropylbenzylamine was tested using the sodium nitroprusside field test kit, a blue color identical to that of methamphetamine resulted [5]. Instrumental testing confirmed that methamphetamine was not present. Page 11 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Isopropylbenzylamine and methamphetamine are isomers, meaning they have similar structures (Figure 1). It is intuitive that the two substances would give similar color results because they both contain ring structures with secondary amines. The LVMPD Forensic Laboratory’s research performed in 2008 showed that a portable Raman device can differentiate between the two extremely similar structures. Figure 1. Structures of methamphetamine and isopropylbenzylamine. Methamphetamine [10] Isopropylbenzylamine [11] Diphenhydramine, commonly known by the trade name Benadryl, and lidocaine, an anesthetic, have been discovered to give positive results for the Scott’s color test when tested by officers in the field. This test is used to presumptively identify cocaine [12], with a two-layer, pink over blue result being positive. There have been appearances of these non-controlled substances in various LVMPD submissions purported to contain cocaine. The suggested partial mechanism of how the first step of the Scott’s test works, and how its blue color is developed, is that the cobalt within the reagent coordinates with the nitrogen on the molecule [42]. Even though diphenhydramine, lidocaine, and cocaine each contain a tertiary amine group, they do not share such strong structural similarities like methamphetamine and isopropylbenzylamine. Knowing that structures of cocaine, lidocaine, and diphenhydramine are so different (Figure 2), and that they are all capable of reacting with the Scott’s test, it is logical to use an instrumental preliminary testing method that differentiates substances based on those structural dissimilarities. Page 12 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 2. Structures of cocaine, lidocaine, and diphenhydramine. Cocaine [10] Lidocaine [10] Diphenhydramine [10] Most forensic laboratories utilize mass spectrometry (MS) and Fourier transform infrared (FTIR) spectroscopy as methods of conclusive identification. Both of these technologies identify compounds based on their molecular structure [13]. The results produced are typically unique to one molecule, especially when coupled with gas or liquid chromatography (GC or LC). Unfortunately, both MS and FTIR have limitations that make them less than ideal candidates for conversion to very small, portable field devices. Many plastics absorb IR light and can cause difficulties with collecting the desired spectrum [37]. Rather than collecting absorbance data, like that of IR spectroscopy, Raman spectroscopy relies on the scattering of light. In Raman processes, a laser excites the bonds in a compound causing changes in molecular vibrational energy which inelastically scatters the light from the laser. When the light is scattered, a new energy response is emitted. It is this response of scattered light that essentially generates the information used to create a spectrum specific to a molecule. [14, 15] Since many illicit drug packages are transparent, Raman spectra can be collected directly through the package making contamination and degradation of the sample unlikely [37]. Like MS and FTIR, Raman spectroscopy has the ability to identify a substance based on molecular structure, but unlike the other instrumentation, it is easily converted into a small handheld system that can test through plastic packaging. Suspected methamphetamine and cocaine are frequently encountered and tested using three separate presumptive field test kits. Being able to eliminate three color test reagents from the field testing program would aid in cost savings and supply issues. Both drugs have been evaluated using various Raman-based instrumentation [3, 25, 35, 43, 44]. Two similar studies indicated that portable Raman field testing was less than 75% accurate in reporting correct results, which leads to the supposition that improvements are needed. Radtke and Patterson, from the Minnesota Bureau of Page 13 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Criminal Apprehension (MBCA), completed a validation study on mixtures of methamphetamine and mixtures of cocaine using a handheld Raman device called FirstDefender manufactured by Ahura (now Thermo Scientific). When they tested case samples, 27.6% of their results were found to be incorrect when compared to GC/MS results (these were mostly false negatives). [27] The National Forensic Science Technology Center completed an evaluation of the FirstDefender RM (Thermo Scientific) and found the device to be approximately 61-69% accurate across various sample sets of various substances using the “point-and-shoot” method. The retail price of this Raman device at the time of the latter discussed study was approximately $50,000. [43] Not only should the accuracy be higher if it is to be implemented in a field testing program, but the system should be more affordable and easily implemented. Portable Raman devices have a few limitations when compared to the more sophisticated laboratory-grade spectrometers and microscopes [4]. Fluorescence, which can cause a high background that can essentially mask the desired spectrum, is one limitation that has been addressed in sophisticated instrumentation, but has not been remedied in portable equipment [16]. Fluorescence in Raman spectra is often encountered when testing dark colored substances [4, 36] or compounds where fluorescence is inherent [16]. There are concerns posed by the LVMPD Forensic Laboratory and other researchers about diluents and adulterants that are often mixed with illicit substances which may fluoresce under certain light excitations. Fluorescence has been an issue with cocaine hydrochloride mixtures seen in evidence submissions [3, 16]. Three studies reviewed were able to identify cocaine in mixtures quantitatively using bench-top Raman spectrometers [3, 25, 26]. Littleford and colleagues described successful analysis of 75% and 86% pure cocaine; however, they do not believe they would have had as much success using field instrumentation without fluorescence rejection [26]. Ryder has also completed quantitative studies on narcotic mixtures with laboratory-grade Raman equipment. He discussed using the “most intense peaks” in narcotic spectra for “rapid discrimination” of the different classes of drugs examined [25]. Using this information, the portable Raman devices may show promise in detecting cocaine in mixtures with future enhancements, such as automated baseline correction and peak detection [30, 41]. Researchers have had success testing opiates such as codeine, morphine, and hydrocodone, which are known to be poor Raman scatterers [31]. However, this was done using Surface Enhanced Raman Scattering (SERS) which is not beneficial to this research. Although reference material Page 14 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 samples of heroin hydrochloride present viable spectra at 785nm [3, 35], previously documented success of testing black tar heroin using confirmatory Raman spectroscopy in a research setting has not been located. Due to the tar’s dark brown-black color and other chemical interferences, the fluorescence is completely overpowering, and sample burning is certain. The TruNarc, a handheld Raman device produced by Thermo Fisher Scientific, has demonstrated success in presumptively testing black tar heroin using a patented sampling system that involves diluting the substance in ethanol and then evaporating the liquid on a metal wafer sampling device [40]. Using graphene as a substrate to quench fluorescence in Raman analysis has also been successful and may be useful for testing a substance such as black tar heroin [45, 46]. Ecstasy tablets pose additional problems with Raman testing as they are often mixtures of several different compounds, come in a variety of colors, and could contain multiple binding agents. These not only cause fluorescence issues, but mixture interferences as well. There are published accounts of using Raman microscopy to “map” ecstasy tablets, but this is both time consuming and not applicable to portable technology [33]. Analysis of plant materials using portable Raman technology and some bench-top instruments has been regarded as difficult or impossible by researchers in the past. Many researchers had to chemically prepare the plant substance in order to test it with Raman instrumentation [23], however, with recent advances, plant material may be analyzed non-destructively [18-24]. Gierlinger and Schwanninger state that their research demonstrated that near infrared Fourier transform (NIR-FT) Raman eliminates fluorescence which makes analysis of green plant material simple [23]. Another set of researchers were able to non-destructively analyze components in a cashew nut without chemically altering the plant material [19]. Although fluorescence interference is encountered in Raman analysis, whether portable or not, it may be combatted utilizing certain specialized fluorescence rejection techniques and technologies which are typically more applicable to laboratory grade instrumentation. The development of SERS alleviated some fluorescence issues in Raman analysis of controlled substances [26, 31], but it involves the use of a Raman microscope and chemicals, which is not applicable to a field test program. Researchers have found that using longer wavelengths and near infrared Raman excitation reduces fluorescence, but this increases the likelihood of sample burning [26, 32, 35, 38], and adds complexity to the instrumentation [38]. Kerr gated Raman spectrometers have proven effective in rejecting fluorescence from street samples of cocaine and cutting agents. This type of analysis Page 15 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 requires special hardware attached to a laboratory based Raman system and high powered lasers which precludes it from being utilized by a portable Raman device [26]. SERDS (shifted-excitation Raman difference spectroscopy) has shown to be useful in the analysis of fluorescing compounds and research has been completed with a focus of translating this technology to a portable system. It requires the implementation of two laser diodes with slightly off-set fixed wavelengths. This technology can be extended to portable Raman spectrometers, but is enabled by coupling an instrument to a separate laser source. [38,39] Chemometric baseline correction methods have been evaluated by various researchers and are typically applied to portable Raman systems due to the capability of these methods to be automated. The ReporteR, manufactured by SciAps, Inc. (previously DeltaNu, Inc.), uses a derivative method of baseline correction. Derivative based correction methods have shown to be useful in reducing noise and variations in background features, but are not as beneficial in resolving high baselines [34, 35] as are often seen in fluorescence laden spectra. Polynomial baseline correction has also been used by researchers to correct fluorescence in Raman spectra, but this cannot be applied automatically without first determining the degree of the polynomial needed [30, 35, 34]. This approach is not feasible in a police officer’s field testing situation. A viable option which may translate to a portable Raman device is the incorporation of automated fluorescence correction technology. Bruker, Inc. manufactures the Senterra Raman microscope which employs a semi-automated baseline correction method in its OPUS software. This company’s technology, known as ‘convex function f’ baseline correction, utilizes a “rubber band” convex polynomial that fits along the baseline of the spectrum in order to find and subtract the “convex envelope” below the spectrum and correct the baseline. The main advantage to this specific technology is that it is applied as an algorithm and was particularly created to correct baselines caused by fluorescence. [30] A patent was published in January of 2013 by Wright, et al, entitled “Methods of Automated Spectral Peak Detection and Quantification Without User Input” that provides a method of automatically applying baseline correction [41]. This invention lends hope that future technology has the potential of not requiring any manipulation on the user end to correct spectral interferences and therefore may be translated to an officer friendly portable system, unlike other fluorescence rejection techniques and technologies. 3. Rationale for the Research By testing the handheld devices in the lab, and in the field by narcotics officers, the LVMPD can determine what features about the device need to be enhanced to increase its user friendly Page 16 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 attributes, efficiency, selectivity, and reliability. This also aids in developing a successful field testing and training program. Whether or not the officers will be able to accept and use the technology successfully is evaluated by this study. This portion of the research also compares the accuracy of the portable Raman handheld device versus the currently employed chemical color test field kits. Higher accuracy, in terms of fewer false positives, as compared to the color test kits is expected for the Raman technology due to the fact that the instrumentation should provide results based on the whole molecular structure versus only a functional group on that molecule. For methamphetamine and cocaine, a parallel study of the spectra collected by the handheld device versus the Raman microscope will provide the data necessary to identify specific spectral variances that are unique to the substances being tested and lead to further enhancement of the portable Raman technology. Not only are the fluorescence correction capabilities valuable to the study of the cocaine mixtures, the confirmatory instrumentation will aid in identifying more specific spectral features to that of cocaine, versus the presumptive, handheld device. The microscope’s accuracy and resolution can discern the subtleties that the handheld units may not detect. It is understood that current portable Raman technology may not be able to directly measure marijuana, ecstasy, or black tar heroin as powerful fluorescence rejection tools are not normally applied or available with portable testing systems. A Raman microscope with an automated or semi-automated fluorescence correction option is essential in this research process to determine if fluorescence correction technology is effective for cocaine mixtures, black tar heroin, ecstasy, and marijuana. If repeatable spectra are collected utilizing the Raman microscope for substances where fluorescence precluded their identification previously, then confirmed against traceable reference materials, the application of this fluorescence correction technology to a portable system in the future is feasible. Based on the research findings, suggestions were made to the manufacturer that enhanced the current portable technology. The LVMPD Forensic Laboratory ultimately hopes that these studies will help establish case law that may assist other local and state districts to utilize this technology in the field. This scientific evaluation and justification for the court system is of utmost importance. Most police officers do not have the expertise to evaluate spectra and to explain to the trier of fact how the technology works. The handheld device can give an officer the results he needs to present in Page 17 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 court, but the court must first rule that it is a valid and reliable tool in preliminary identification of controlled substances (satisfy the Frye standard). The development of a successful field testing program that will replace the use of chemical field test kits is the ultimate goal of this research. To determine if testing typically difficult samples with the fluorescence baseline correction algorithm is achievable, and to explore the possibilities of applying this technology to a handheld device is an important portion of the research. The information obtained from the studies completed will aid in creating a definitive, new method of field testing that will enhance the ability of the law enforcement community to presumptively identify controlled substance evidence. II. Methods Beginning in 2008, preliminary research had been completed in conjunction with DeltaNu, Inc. (now SciAps, Inc.), the manufacturer of a handheld Raman device called the ReporteR [28]. While working with DeltaNu, Inc., a small number of improvements were made to the firmware and design of the ReporteR. These changes took place prior to the National Institute of Justice award for this project. After applying for this grant, but prior to initiating this study, further enhancements to the ReporteR were completed by the manufacturer, including the addition of a “test number” to be displayed on the screen. The “test number” was intended to facilitate coordination between the evidence case number and the data stored on the device so that a spectrum can be viewed at a later time. It was discovered during data analysis that the test number did not correlate to the data file. This was remedied during the third phase of testing after enhancements had been completed. After the LVMPD was awarded this grant, six ReporteRs, chemical color test field kits, and a Bruker Senterra Raman microscope were purchased (the Raman microscope was purchased after one year of research was completed). Upon receipt, the ReporteR devices were calibration checked and the color test field kits were quality control checked according to LVMPD Forensic Laboratory procedures, using traceable reference materials. All primary, secondary, non-controlled and controlled reference materials used in the study were provided by LVMPD. The verified reference materials used throughout the study are included in Table 3. During analysis, all appropriate blanks and quality control checks were completed in accordance with the laboratory’s requirements and procedures. Page 18 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 A paper checklist for methamphetamine and cocaine was developed for officers so that the required procedure would be followed when performing the Raman handheld test. This ensured that the tests were consistently and correctly performed in the field. The checklist that was developed is shown in Figure 3. Throughout the testing process, various LVMPD lab-created libraries were loaded onto the handheld devices and used by both the field officers and in the laboratory. This was done to determine the best combination of reference spectra for the most accurate testing. The reference materials included in the libraries are shown in Tables 1 and 2. The ReporteRs were validated by testing known controlled and non-controlled reference materials against the libraries that were created using the accompanying software. In addition, 37 neat negative controls, similar in appearance and/or structure to methamphetamine and cocaine, and negative control mixtures were examined to study the potential of receiving false positive results and to test the specificity and selectivity of the handheld unit by testing each with a ReporteR through plastic bags. They were also tested with color test field kits. Positive controls of confirmed casework samples were also run against the created libraries (Tables 4.1-4.7). Figure 3. Field Checklist. Page 19 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Two different ways of displaying results were tested on different versions of the ReporteR with the changes implemented by DeltaNu, Inc. (SciAps, Inc.). One version of the results display screen included the listing of the actual library reference material name and the actual correlation value (e.g. “dimethylsulfone 0.69”). The other version included the listing of “methamphetamine,” “cocaine,” or “no match” and a bar value which was set to a certain correlation. See Figure 4. The number of bars and corresponding correlation values were concurrently tested while trying to assess the regular incidence of false negative results with cocaine mixtures. Approximately 40% and 50% cocaine/caffeine mixtures (by weight) were used to test different levels of bar correlations with 5 bars being the best match, ranging from 0.60 to 0.85. Negative controls were tested against the 5 level bar correlation which was set to 0.80 to 0.90. For all correlation settings, see Table 6. Figure 4. Photos of display screens. In the first phase of methamphetamine and cocaine testing, samples were analyzed solely by the laboratory. Each forensic scientist participating in the research created a database describing the evidence and its packaging, the case number, the item number, the inventory number of the ReporteR used, and the results of each test performed. Research testing continued until forensic scientists had tested over 280 white to off-white powders, crystalline substances, and chunky substances from adjudicated and current casework evidence samples with the ReporteR and chemical field tests. This was a side-by-side study of the current field testing method with the proposed field testing method. This study tested the capabilities of the instrumentation and whether or not it is as specific and as sensitive as the chemical field test. When a spectrum was obtained, it was saved electronically and/or as a hardcopy. While testing the casework samples, each forensic scientist performed varied calibration checks on the ReporteR and recorded the Page 20 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 results in their database. This was completed in order to determine the length of a valid calibration and test the reliability of the device. The evidence samples were confirmed by gas chromatography/mass spectrometry (GC/MS) (see parameters included at the end of this section). In order to test the reliability between devices, items of controlled substance evidence from some of the adjudicated cases mentioned above were retested randomly. Adjudicated cases to be retested were selected at random after initial testing, so a few items were retested by the same forensic scientist. In the second phase of methamphetamine and cocaine testing, narcotics detectives from varied assignments (LVMPD Narcotics, an outside jurisdiction, an interdiction team, and a federal task force) were trained on the use of the ReporteR, checklists, and procedures. Detectives completed testing on 140 white to off-white powder or crystalline case samples in the field using a ReporteR and chemical field test kits. The detectives completed the corresponding checklists for both chemical and Raman tests. The checklists were included in the evidence package and delivered to the laboratory through LVMPD evidence handling procedures. The data from the handheld devices was stored on a disk inside the units and was periodically collected by the laboratory. After the handheld devices were enhanced to include portable Bluetooth printers, the printouts (Figure 5) replaced the Raman checklist and were included with the evidence. All field tested evidence was retested in the laboratory using ReporteRs and confirmatory testing procedures used by the LVMPD (the compounds in the substances were confirmed via GC/MS). It was determined that, as the prevalence of isopropylbenzylamine had waned, and the 100% accuracy of first phase color testing, methamphetamine chemical field test kits would no longer be performed in the remaining laboratory portion of the testing (but would continue in the field) unless a false positive field result was discovered. Cocaine field color test kits would be used both in the field and the lab. Two different sampling attachments were tested to discover whether or not a specific design was more amenable to testing through plastic bags. In preparation for the field testing, the forensic scientists attached a chain to the device to secure the sampling attachment to prevent loss. As manufactured, the attachment fits onto the device and is held by a weak magnet. The attachment was easily removed and had the potential to become separated from the device. Page 21 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 5. Photo of Bluetooth printer and printout. Three methods of polystyrene “self-testing” were evaluated in the field. 1) A built in calibration check “lock” that forces a tester to run a polystyrene check before each test. 2) Removal of any forced self-test and relying on the officer to complete the self-test on a regular basis. 3) A forced self-test in which an administrator can set the length of time allowed between self-tests. For both the laboratory and field research, the handheld device testing procedure consisted of testing each item, at a maximum, three different times at three different locations on the item. Possible result values were methamphetamine, cocaine, (or another drug in the library depending on library version), “No Signal,” or “No Match.” A “No Signal” result did not equate to a “No Match” result. If a tester received a “No Signal” result, the test was repeated. The assumption is that if at least one of the three tested locations contains methamphetamine or cocaine, the LCD screen will indicate that result. Testing was performed through its original packaging when practical. If the original drug packaging was unsuitable for Raman testing, the substance was placed into a 2 mil polyethylene bag. Page 22 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 The triplicate testing procedure was realized from discussions with narcotics detectives and from testing completed prior to grant funding. Due to all the responsibilities officers have, they require a simple procedure to follow during a testing process. The more complicated a procedure is, the more likely errors will be made or they will choose not to perform the testing at all. Laboratory personnel determined that a maximum of three tests should be adequate to determine if a target controlled substance is present in the sample. If it is not detected after three tests, then it is possible that interfering diluents in the sample are overwhelming the target analyte, preventing it from being detected by the ReporteR and requiring laboratory analysis. Allowing a large or unlimited amount of testing could increase the number of erroneous results. The purity of the substance is not a concern for this study because the homogeneity of samples will vary, and the chance of testing a non-target crystal is possible regardless of purity. Since the focus of this study is in determining the effectiveness of this technology as a field testing tool, and there is no way to determine purity on the street, the quantitation of the study samples was not deemed a necessary component of this research. The third phase of methamphetamine and cocaine testing included the incorporation of the Raman microscope and the use of an enhanced ReporteR in the field. After the devices were upgraded and printers added, further training was provided to the narcotics detectives. Each evidence sample was subjected to field testing with both color test panels (3 kits/2tests) and a ReporteR, and laboratory testing with a ReporteR, color test kits (cocaine only), and the Raman microscope (see parameters at the end of this section). After installation of the Raman microscope, 26 substances were placed into plastic bags, analyzed using the Bruker Senterra, and loaded into a lab-created library. Twenty negative reference material controls and 13 positive controls consisting of confirmed casework samples and mixtures of cocaine and diluent reference materials were tested against the library. A comparison of selected methamphetamine and cocaine spectra collected in the field and laboratory was completed using the OPUS software (Bruker, Inc.). An evaluation of cocaine versus commonly accompanying diluents was completed using the Raman microscope. Peak location determinations and baseline corrections were completed using the OPUS software. At the time of writing, over 560 case work samples were included the study. The majority of testing data is included in this final report, however, either due to incomplete testing (no chemical field test performed, no Raman results found, etc.), instrument malfunctions, protocol deviations, Page 23 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 or as of the time of writing, the case could not be analyzed by the laboratory due to time constraints, some data was not evaluated. After a significant amount of testing, the LVMPD Forensic Laboratory requested that the manufacturer of the ReporteR make specific software and firmware changes to the device to further enhance its usability. The final enhancements included: 1) A small portable Bluetooth printer that produces a customizable printout which includes the serial number of the device, the data file, the last self-test date and result, the sample test result and correlation, and blank fields for handwritten entry to include case information and signature. This documentation eliminates the need for a checklist. The “test number” was replaced by the data file name on the print out. The results are traceable back to the data saved on the SD card. 2) Enhanced software that has an improved signalto-noise ratio. 3) A new polystyrene standard that prevents an officer from having to switch attachments in order to perform a self-test. 4) A forced self-test in which an administrator can set the length of time allowed between self-tests. 5) A stronger magnet that secures the removable sampling attachment to the device. Marijuana ReporteR Testing Prior to receipt of the Raman microscope, marijuana and delta-9-tetrahydrocannabinol testing was attempted with the ReporteR. It was unknown at that time whether a portable Raman liquid sampling system that incorporates an extraction method could be developed in order to test plant material in the field, as testing it directly is not feasible with Raman-based portable technology. However, using a liquid extraction method does not support the idea of eliminating the use of chemicals in the field. Not only is fluorescence an issue when testing plant material using Raman spectroscopy, but the fact that plants, by nature, contain a variety of chemical compounds which also may result in interference. A sampling system would have to include the simplest and least expensive materials in order to fit the goal of creating an inexpensive field testing system (the ReporteR is equipped with a removable vial holder attachment in order to test liquids). Several methodologies of isolating cannabinoids from plant material and from an ethanolic reference material were tested using the ReporteR and accompanying NuSpec software. This testing included using various solvents and solutions (evaporated onto slides and dissolved in liquid sampling devices), filter paper, thin layer chromatography (TLC), and water based extractions. Materials included delta-9-tetrahydrocannabinol (in ethanol, 25 mg/mL, Sigma) , secondary marijuana reference material, white filter paper (Whatman, Grade 3, 6µm), silica gel Page 24 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 GHLF 250 micron plate (Analtech), 1.0 N NaOH (Fisher, concentration made in house), 2.0 N HCl (BDH, concentration made in house), methylene chloride (JT Baker), distilled water (Sparklett’s), isobutanol (Shelton), pentane (Aldrich), acetone (EMD), methyl ethyl ketone (Sigma), hexanes (EM Science), acetonitrile (EMD), isopropanol (Honeywell B&J), ethanol (Sigma), methanol (Honeywell B&J), ethyl ether (BDH), petroleum ether (EMD), benzyl alcohol (Fisher). The blank solvents and solutions listed above were tested to confirm their Raman signatures. The liquid sample vials and attachment provided with the ReporteR were used. Water, 1.0 N NaOH, 2.0 N HCl, and methylene chloride were selected for testing and the extracts were tested via the liquid sampling attachment. Because most common solvents exhibit strong Raman spectra (as was confirmed by blank solvent testing using the ReporteR), solvents either need to be evaporated to avoid interference, or the solvated compound needs to overwhelm the solvent signal. During initial attempts, as the reference material was ethanolic, the delta-9-tetrahydrocannabinol was spotted onto filter paper. The concentrated spot was isolated, soaked, and vortexed in distilled water. The water was placed into a sampling vial and tested using ReporteR and the liquid vial attachment. The delta-9-tetrahydrocannabinol reference material and a methylene chloride extracted marijuana reference material were spotted onto a TLC plate and run in Hexane/Acetone 3:1 mobile phase. A blank spectrum of the TLC plate was gathered prior to spotting. Spectra of both the initial reference and sample spots were taken. The plate was visualized with UV light. The marijuana solvent front spot with a corresponding Rf value to the delta-9-tetrahydrocannabinol reference material was tested with the ReporteR and NuSpec software. Attempts were made to collect spectra for marijuana and negative secondary reference plant materials by way of direct testing using the ReporteR and the accompanying software. After testing with the Raman microscope had commenced, an evaporated sample of delta-9tetrahydrocannabinol on an aluminum slide was tested using the ReporteR. Marijuana Microscope Testing The following reference materials were examined using the Senterra Raman microscope and accompanying OPUS software: delta-9-tetrahydrocannabinol (Sigma), delta-8tetrahydrocannabinol (Sigma), cannabinol (Sigma), cannabidiol (Grace), and cannabichromene (Restek). Reference materials that were in liquid form were evaporated onto slides. Both 532 nm Page 25 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 (10 mW, 400 grating) and 785nm (100 mW, 400grating) lasers were tested at various integration times and co-additions. Glass, gold, silver, and aluminum slides were evaluated. Marijuana plant material case work samples from ten different events, and two suspected marijuana “wax” samples were tested with the Senterra Raman microscope at 785 nm on aluminum foil slides. Evidence samples were confirmed via stereoscope (plant material only), GC/MS, and Duquenois-Levine (ethanol (Mallinckrodt, Acros), acetaldehyde (Alfa Aesar), vanillin (Fisher, Alfa Aesar), and chloroform (J.T. Baker)) color testing (same procedure as the chemical field tests). Secondary negative plant materials were also examined using the Raman microscope. Photos were taken using the microscope’s OPUS software for each of the plant materials illustrating what each looked like compared to marijuana. Only those plant materials that exhibited resinous structures similar to marijuana were tested using the Raman portion of the instrument utilizing aluminum slides at 785 nm (100 mW, 400 grating, and 5 second integration time with 2 coadditions). Black Tar Heroin Testing This evaluation was included in the study because a large portion of drug evidence seen in the immediate area consists of black tar heroin. As with marijuana, testing black tar heroin directly is not feasible with Raman-based portable technology. As other manufacturers had developed a sampling system for heroin, the researchers were curious as to whether they could develop ideas for a portable Raman sampling system. Various methods were tested using both the microscope and the ReporteR. Casework samples of black tar heroin were used. The substance was run neat on aluminum foil slides, solvated with water and ethanol and evaporated onto gold and aluminum slides, and then tested using the microscope. A mixture of black tar heroin and potassium bromide (equal parts) was tested through a plastic bag with both instruments. Mixing of the substance with graphene (Angstron Materials) by way of liquid sampling and concentrating onto glass fiber filters was also evaluated. A small amount of black tar heroin (~1 mg) was solubilized in water (~1 mL) with a small amount of graphene (~1 mg), vortexed and then evaporated dropwise onto the glass fiber filters. The concentrated spot was allowed to dry in ambient laboratory conditions at room temperature over the course of 1 day and then again over 2 weeks. The 785 nm (100 mW, 400 grating) laser was used. Spectra collected from the ReporteR and the Page 26 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 microscope were compared to each other and to a heroin reference material using the OPUS software. Heroin was confirmed via GC/MS. Ecstasy Testing Ecstasy tablets are also routinely encountered by the LVMPD Forensic Laboratory. Contrary to marijuana and black tar heroin, testing ecstasy tablets directly is possible with Raman-based portable technology, but limited. It was of particular interest as to whether baseline correction technology would assist in the analysis of such drug evidence. Eight ecstasy samples were tested using the Raman microscope at 785 nm (100 mW, 400 grating). Microscopic photos using the 20x objective were taken; however, the color of the tablets was not apparent microscopically. Two of these samples were also tested using the ReporteR. The samples were tested without preparation through plastic bags. The evidence varied in color and was confirmed via GC/MS. Emergent Synthetic Drug Testing Even though they were not included in the original proposal, various synthetic drugs were also tested as the recent surge of these compounds has caused certain concern in the law enforcement and health community. The LVMPD Forensic Laboratory had interest in determining which compounds could be detected using Raman spectroscopy and if baseline correction would aid in their identification. Certain available reference materials were tested using the Raman microscope at 785 nm (100 mW, 400 grating) and/or the ReporteR. A few casework samples, including “spice” plant material, were also tested. GC/MS parameters The confirmatory analysis was conducted in scan mode using an Agilent 7890A gas chromatograph coupled to an Agilent 5975C mass spectrometer with a mass scan range of 39 to 500 mass to charge ratio, an electron energy of 70eV, and ion source temperature at 250°C. The GC columns used were DB-5MS (Agilent J&W) at either 20 m long with a 0.18 mm diameter and 0.18 µm film thickness (hydrogen (H2) carrier gas), or 30 m or 15 m in length with a 0.15 mm diameter and 0.15 µm film thickness (helium (He) carrier gas). The injection volume was either 0.5 µL (H2) or 1.0 µL (He) on split (10:1) injection. The inlet was set to 215º C (H2) or 250 º C (He) with a flow of 0.85 mL/min (H2) or 1 mL/min (He). Ramping temperature programs varied: H2: 75º C (0.452 min), 32º C/min to 215º C (0 min), 25º C/min to 200º C (0 min), 32º C/min to 280º C (9.573 min); 75º C (0.452 min), 32º C/min to 215º C (0 min), 25º C/min to 200º C (0 Page 27 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 min), 33º C/min to 280º C (2.1 min); 170º C (0.5 min), 30º C/min to 275º C (0 min), 1º C/min to 274º C (1 min), 30º C/min to 295º C (1.5 min). He: generally 75º C ramped at 25-30º C/min to 280º C (8 min). Samples were dissolved or extracted using appropriate solvents and solutions (methanol (Honeywell B&J), chloroform (JT Baker), methylene chloride (JT Baker), water/sodium hydroxide (Fisher)/organic solvent (chloroform or methylene chloride, JT Baker)). Raman parameters Comparative analysis of methamphetamine and cocaine, and other research was conducted using a Bruker Senterra Raman microscope equipped with 785nm and 532 nm lasers, 400 rulings/mm holographic grating (9-18 cm-1 resolution), a CCD detector, power at the sample ranging from 10 mW to 100 mW, an Olympus microscope (20x objective used) and OPUS software (with integrated baseline correction function). Presumptive Raman analysis was completed using DeltaNu, Inc., (SciAps, Inc.) ReporteRs, V2, with a 785 nm laser at ~12-15 cm-1 resolution, with 120~50 mW power at the sample, a silicon CCD 2048 liner array 1800 grating a laser spot size <50 micron, and NuSpec software. The microscope has a higher laser power density and a smaller aperture that will increase the resolution. The microscope has a set number of scans (5), but the ReporteR’s number of scans may vary during collection. The microscope has two available gratings and the ReporteR only one at 1800, but the resolutions are approximately the same (9-18 cm-1 vs. ~12-15 cm-1) when using the Senterra’s 400 grating. During select analyses, metallic slides (gold, silver, aluminum) were used. All spectra have x-axes in wavenumbers (cm-1 ) and y-axes in Raman intensity. Statistical Analysis Accuracy of results is reported as percentages of the total number of data points evaluated in the data analysis. As the scope of this research is to ultimately serve the law enforcement community, percentages are an effective way to communicate results so that a broader public can benefit from the information generated by the study. Throughout the report, “accuracy rate” is used to describe results. For example, the accuracy rate of the ReporteR was determined by whether or not it correctly detected the target analyte in the sample. Did it detect cocaine when cocaine was present? Did it result in “no match” when no cocaine or methamphetamine was in the sample? This was then recorded as a percentage of the entire data set. Page 28 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 1. Microscope Library Acetaminophen (Sigma) Acetylsalicylic Acid (Kodak) d,l-Amphetamine (Sigma) Benzocaine (Sigma) Benzoylecgonine (Sigma) Caffeine (Sigma) Cocaine base (Sigma) Cocaine HCl (Sigma) (+/-) Deoxyephedrine (Sigma) Dimethylsulfone (Sigma) Diphenhydramine (Parke Davis) (-) Ephedrine (Sigma) (+/-) Ephedrine (Sigma) Heroin (Sigma) Isopropylbenzylamine (IPBA) (Alfa Aesar-crystallized in house) Ketamine (Sigma) α-Lactose (Sigma) Levamisole (MP Bio) Lidocaine (unknown, verified) Mannitol (Rim Labs) Methamphetamine (Sigma) Phentermine (Sigma) Phenylephrine (Sigma) Procaine (Sigma) Pseudoephedrine (Sigma) Tetramisole (Sigma) Table 2. ReporteR Libraries First Library Version d,l-Amphetamine (K&K Labs) Polystyrene Calibration Standard (DeltaNu, Inc.) Cocaine base (Sigma) Cocaine HCl (Sigma) Table 2. ReporteR Libraries continued Second Library Version d,l-Amphetamine (K&K Labs) Polystyrene Calibration Standard Cocaine base (Sigma) Cocaine HCl (Sigma) Dimethylsulfone (Sigma) (+/-) Ephedrine (Sigma) Heroin (Sigma) Ibuprofen (Sigma) IPBA (Alfa Aesar-crystallized in house) Levamisole (MP Bio) Lidocaine (unknown, verified) Methamphetamine (Sigma-2 lot numbers) Procaine (Sigma) Pseudoephedrine (Sigma) Third Library Version d,l-Amphetamine (K&K Labs) Polystyrene Calibration Standard Cocaine base (Sigma) Cocaine HCl (Sigma) Dimethylsulfone (Sigma) Diphenhydramine (Parke Davis) (+/-) Ephedrine (Sigma) Heroin (Sigma) Ibuprofen (Sigma) IPBA (Alfa Aesar-crystallized in house) Ketamine (Sigma) Levamisole (MP Bio) Lidocaine (unknown, verified) Methamphetamine (Sigma-2 lot numbers) Pseudoephedrine (Sigma) “Crack” Cocaine Base/Sodium Bicarbonate (EMD) –“cooked” in house, equal parts Dimethylsulfone (Sigma) (+/-) Ephedrine (Sigma) Ibuprofen (Sigma) IPBA (Alfa Aesar-crystallized in house) Levamisole (MP Bio) Lidocaine (unknown, verified) Methamphetamine (Sigma-2 lot numbers) Procaine (Sigma) Pseudoephedrine (Sigma) Page 29 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 3. Methamphetamine & Cocaine Negative and Positive Controls and Library Validation ReporteR Testing Acetaminophen (Sigma) Acetylsalicylic Acid (Sigma) d,l-Amphetamine (K&K Labs) Benzocaine (Sigma) Benzoylecgonine (Sigma) Caffeine (Sigma) Cathine (NMI) Cyclobenzaprine (Sigma) Deoxyephedrine (Methamphetamine) (Sigma-different lot number than in library) Dextromethorphan (Robins Research) Dimethylsulfone (Sigma) Diphenhydramine (Parke Davis) (+) Ephedrine (Sigma) (-) Ephedrine (Sigma) (+/-) Ephedrine (Sigma) Ethylbenzylamine (Alfa Aesar-crystallized in house) Methylenedioxyethylamphetamine (NMI) Ibuprofen (Sigma) IPBA (Alfa Aesar-crystallized in house) JWH-018 (Fluka) Ketorolac (USP) α-Lactose (Sigma) Levamisole (MP Bio) Lidocaine (unknown, verified) Mannitol (Rim Labs) Methylenedioxyethylamphetamine (NMI) Methylenedioxypyrovalerone (MDPV) (Cayman Chemical) Methylphenidate (USPC) Morphine (USPC) Naproxen (Sigma) Oxycodone (Sigma) Phentermine (Sigma) Phenylpropanolamine (Sigma) Procaine (Sigma) Pseudoephedrine (Sigma) Salicylic Acid (MP Bio) Tetracaine (Sigma) Tetramisole (Sigma) Negative control mix of phentermine and dimethylsulfone Microscope Testing Acetaminophen (Sigma) Acetylsalicylic Acid (Sigma) Amoxicillin (Sigma) d,l-Amphetamine (K&K Labs) Caffeine (Sigma) Cathine (NMI) Cyclobenzaprine (Sigma) Dextromethorphan (Robins Research) (+) Ephedrine (Sigma) Ethylbenzylamine (Alfa Aesar-crystallized in house) Ibuprofen (Sigma) JWH-018 (Fluka) Methylenedioxyethylamphetamine (NMI) Methylphenidate (USPC) Oxycodone (Sigma) Phenylpropanolamine (Sigma) d-Propoxyphene (Applied Science) Pseudoephedrine (Sigma) Salicylic Acid (MP Bio) Tetracaine (Sigma) ~50% Methamphetamine/~25% Isopropylbenzylamine/~25% Dimethylsulfone ~50% Caffeine/~50% Cocaine Hydrochloride “Crack” Cocaine Base/Sodium Bicarbonate (EMD) –“cooked” in house, equal parts Continued on next page Page 30 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 3. Methamphetamine & Cocaine Negative and Positive Controls and Library Validation Synthetic Drug Testing ReporteR Butylone (Cayman Chemical) Methylone (Cayman Chemical) 4-Methylethcathinone (MEC) (Cayman Chemical) MDPV (Cayman Chemical) Methylone (Cayman Chemical) α-Pyrrolidinopentiophenone (PVP) (Cayman Chemical) UR-144 (Cayman Chemical) XLR-11 (Cayman Chemical) Microscope Butylone (Cayman Chemical) o-Chlorophenylpiperazine (Sigma) 2- Fluoroamphetamine (Cayman Chemical) 3- Fluoroamphetamine (Cayman Chemical) 2- Fluoromethamphetamine (Cayman Chemical) 3- Fluoromethamphetamine (Cayman Chemical) 4- Fluoromethamphetamine (Cayman Chemical) 4-Fluoromethcathinone (Cayman Chemical) p-Fluorophenylpiperazine (Cayman Chemical) Methylone (Cayman Chemical) 4-MEC (Cayman Chemical) MDPV (Cayman Chemical) Methylone (Cayman Chemical) α- (PVP) (Cayman Chemical) UR-144 (Cayman Chemical) XLR-11 (Cayman Chemical) Plant Material Negative Controls 2° ReporteR Catnip Eucalyptus Culinary mushrooms “Hydro” commercially produced synthetic marijuana Psilocin mushrooms Tobacco Microscope Cape myrtle Catnip Coltsfoot Comfrey Dandelion Feverfew Green tea Ground basil Honeysuckle Horehound “Hydro” commercially produced synthetic marijuana Jojoba Kenaf Lantana Marjoram Mint Mugwort Nettle Oregano Pineapple guava Sage Thistle Thyme Tobacco Page 31 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 III. Results Due to the amount of testing performed, not all data that was collected is provided in a table format. A majority of the data is summarized in paragraph or bullet point format. As the amount of spectra collected is sizeable, examples of various studies are provided in figures in the corresponding sections. The results are presented in chronological order of testing performed. ReporteR Validation/Libraries/Negative Testing The following substances listed in Tables 4.1 through 4.7 were tested against the ReporteR libraries. Substances used as negative controls were also tested with two ReporteRs and chemical field test kits. Tables 4.1 – 4.7. Library testing results. A negative chemical field test result is defined as one that is not expected for methamphetamine or cocaine. A positive chemical field test result is defined as an expected result or one that may be interpreted as an expected result. A reference material may give three different ReporteR results as testing through plastic bags may provide sampling and signal strength differences. Table 4.1 First Library Version Tests- No Chemical Field Test Performed Reference Material/Sample Tested 1 ReporteR Result Correlation*/# of Bars Cocaine Cocaine 5 bars Amphetamine1 Amphetamine 2 bars No match ---- (not in library) No match ---- (not in library) Amphetamine 5 bars Ephedrine Ephedrine 4 bars Ephedrine (different lot #)1 Ephedrine 4 bars Pseudoephedrine Pseudoephedrine 5 bars Deoxyephedrine1 Methamphetamine 5 bars Morphine No match ---- (not in library) Levamisole1 Levamisole 4 bars Caffeine No match ---- (not in library) 1 Lactose No match ---- (not in library) IPBA1 IPBA 5 bars Dimethylsulfone Dimethylsulfone 5 bars Procaine1 Procaine 3 bars No match ---- (not in library) Mannitol No match ---- (not in library) Pink cocaine case sample1 Cocaine 1 bar Ketorolac 1 MDEA1 1 Amphetamine (different lot #) 1 1 1 1 1 Acetaminophen 1 1 (#X) = how many times sample was tested with same result *5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.80, 2 bars = 0.79-0.70, 1 bar = 0.69-0.60, No match = 0.59-0.00 1, 2 † corresponds to different ReporteR devices positive results subjective according to tester ** NuSpec software only reports correlation values This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Page 32 of 95 Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 4.2 First Library Version Tests- Negative Chemical Field Test Results Reference Material/Sample Tested 1 ReporteR Result Correlation*/# of Bars Caffeine No match (3X) ---- (not in library) Oxycodone1 No match (3X) ---- (not in library) No match (3X) ---- (not in library) Ephedrine (2X) 2 bars Cathine Ephedrine 3 bars Mannitol1 No match (3X) ---- (not in library) 1 Ephedrine 3 bars 1 Ephedrine Ephedrine (2X) 4 bars Methylphenidate1 Ephedrine 2 bars Methylphenidate Calibration Standard 2 bars Methylphenidate1 Ephedrine 1 bars 1 Procaine Procaine 3 bars Procaine1 Procaine 1 bars Procaine 4 bars Ephedrine different lot number) Ephedrine (3X) 4 bars Acetaminophen1 No match (3X) ---- (not in library) Phenylpropanolamine Calibration Standard (3X) 3 bars Levamisole2 Levamisole 2 bars No match (3X) ---- (not in library) No match (3X) ---- (not in library) No match (3X) ---- (not in library) No match (3X) ---- (not in library) No match (3X) ---- (not in library) No match (3X) ---- (not in library) Dimethylsulfone 3 bars Isopropylbenzylamine 2 bars Ibuprofen 2 bars No match (3X) ---- (not in library) Benzocaine No match (3X) ---- (not in library) Tetramisole2 Levamisole 3 bars Lactose No match (3X) --- (not in library) Dextromethorphan2 No match (3X) --- (not in library) Ephedrine 4 bars Pseudoephedrine 3 bars 1 MDPV 2° Cathine1 1 Ephedrine 1 1 Procaine 1( 1 Acetylsalicylic Acid 2 Caffeine2 Salicylic Acid 2 Acetaminophen 2 Tetracaine2 JWH-018 2 Dimethylsulfone2 2 Phenylpropanolamine Ibuprofen 2 Lidocaine2 2 2 2( Ephedrine different lot number) 2 Pseudoephedrine (#X) = how many times sample was tested with same result *5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.80, 2 bars = 0.79-0.70, 1 bar = 0.69-0.60, No match = 0.59-0.00 1, 2 † corresponds to different ReporteR devices positive results subjective according to tester ** NuSpec software only reports correlation values Page 33 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 4.3 First Library Version Tests- “Positive” Marquis (Methamphetamine) Chemical Field Test Results Reference Material/Sample Tested ReporteR Result Correlation*/# of Bars 1 No match ---- 1 Amphetamine Amphetamine 3 bars Amphetamine1 Amphetamine 5 bars Phentermine Amphetamine 2 bars Phentermine1 Ephedrine 1 bars Phentermine Amphetamine 1 bars Diphenhydramine2 Calibration Standard (3X) 3 bars No match (3X) --- (not in library) Amphetamine 1 1 Naproxen 2 Table 4.4 First Library Version Tests- “Positive” Sodium Nitroprusside (Methamphetamine) Chemical Field Test Results Reference Material/Sample Tested ReporteR Result Correlation*/# of Bars MDEA1 No match (3X) ---- (not in library) Isopropylbenzylamine2 Isopropylbenzylamine 4 bars Ethylbenzylamine2 Calibration Standard (3X) 2 bars Table 4.5 First Library Version Tests- “Positive” Scott’s (Cocaine) Chemical Field Test Results Reference Material/Sample Tested 2 Cyclobenzaprine ReporteR Result No match (3X) Correlation*/# of Bars ---- (not in library) Table 4.6 Secondary Library Version Tests- “Positive” Marquis (Methamphetamine) Chemical Field Test Results Reference Material/Sample Tested 1 ReporteR Result Correlation*/# of Bars Phentermine Amphetamine 1 bar Phentermine1 Ephedrine (4X) 1 bar Phentermine/Dimethylsulfone mixture1 Dimethylsulfone (2X) 4 bars 1 Dimethylsulfone (3X) 3 bars Phentermine/Dimethylsulfone mixture (#X) = how many times sample was tested with same result *5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.80, 2 bars = 0.79-0.70, 1 bar = 0.69-0.60, No match = 0.59-0.00 1, 2 † corresponds to different ReporteR devices positive results subjective according to tester ** NuSpec software only reports correlation values Page 34 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Table 4.7 Third Library Version Tests- No Chemical Field Test Performed Reference Material/Sample Tested ReporteR Result 1 Methamphetamine case sample Correlation*/# of Bars Procaine (2X) Not recorded Cocaine case sample Procaine Not recorded Methamphetamine case sample1 Methamphetamine Not recorded Lidocaine case sample Lidocaine Not recorded Cocaine Base/Sodium Bicarbonate1 Cocaine Not recorded 1 Cocaine Not recorded 1 1 Cocaine Base/Sodium Bicarbonate Removed Procaine (mismatches) and one of the methamphetamine spectra from library (poor spectrum) then tested with NuSpec software below to view correlations** Methamphetamine case sample1 Methamphetamine 0.82 Methamphetamine case sample1 Methamphetamine case sample1 (“Foodsaver” bag) Methamphetamine Lidocaine (See the Methamphetamine section Methamphetamine case sample1 (“Foodsaver” bag) Cocaine case sample1 1 Cocaine case sample 0.79 0.43 in Conclusions, page 81) Dimethylsulfone 0.96 Cocaine HCl 0.88 Cocaine base 0.94 (#X) = how many times sample was tested with same result *5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.80, 2 bars = 0.79-0.70, 1 bar = 0.69-0.60, No match = 0.59-0.00 1, 2 † corresponds to different ReporteR devices positive results subjective according to tester ** NuSpec software only reports correlation values First Phase Methamphetamine and Cocaine Testing Results Laboratory Only Testing Methamphetamine Results:  Total samples included in data analysis: 126  Zero false positive ReporteR results  Positive lab chemical field test results vs. positive identification by GC/MS: 126/126 = 100% accuracy rate (2 reported as weak chemical field test results, which were confirmed to contain methamphetamine and dimethylsulfone)  Positive lab ReporteR results vs. positive identification by GC/MS: 116/126 = 92% accuracy rate  58 of the 126 samples were tested previously by field officers using chemical field tests with positive results. The officer’s chemical field test results agreed with chemical field tests performed in the lab by forensic scientists. Page 35 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201  Inconclusive lab chemical field test results vs. positive identification by GC/MS: 0/126 = 0% false negatives  No match/inconclusive lab ReporteR results vs. positive identification by GC/MS: 10/126 = 7.9% false negatives  No match/inconclusive lab ReporteR results vs. negative identification by GC/MS: 10/0 = 0% (All of the samples chosen for testing were purported to contain methamphetamine, which was confirmed through GC/MS analysis.) Laboratory Only Testing Cocaine/Suspected Cocaine Results:  Total samples included in data analysis: 141  Zero false positive ReporteR results  Positive lab chemical field test results vs. positive identification by GC/MS: 76/104 = 73% accuracy rate (2 reported as weak chemical field test results which were confirmed to contain cocaine)  Positive lab ReporteR results vs. positive identification by GC/MS : 76/104 = 73% accuracy rate  Inconclusive lab chemical field test results vs. positive identification by GC/MS: 28/104 = 26% false negatives  No match/inconclusive lab ReporteR results vs. positive identification by GC/MS: 28/104 = 26% false negatives  No match/inconclusive lab ReporteR results vs. negative identification by GC/MS: 37/37 = 100% accuracy in reporting negative results  77 of the 104 positive cocaine casework samples were originally field tested by field officers and then by forensic scientists during their research. Several discrepancies were encountered between officer’s results and chemical field test results obtained by forensic scientists in the lab illustrating the subjectivity in the interpretation of chemical field tests (see Table 5). Table 5. Comparison of 77 total suspected cocaine items tested with chemical field test kits in the field and in the laboratory. The true nature of the discrepancies between officer and laboratory testing is unknown. These results demonstrate that the officers and laboratory personnel either perform the test differently or interpret the results differently, and that an Page 36 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 objective field test is needed for suspected cocaine samples. See page 82 for explanation of cocaine field test errors. 77 Officer and Laboratory Performed Chemical Field Tests # of officer positive cocaine chemical field test results # of lab positive cocaine chemical field test results 67 48 # of officer false positive cocaine chemical field test results (of 67) # of lab false positive cocaine chemical field test results (of 48) 3 0 # of officer inconclusive cocaine chemical field test results # of lab inconclusive cocaine chemical field test results 10 29 ReporteR Calibration Stability Laboratory Testing Each forensic scientist performed calibration checks using the provided polystyrene standard on each ReporteR unit concurrent with the first phase of casework sample testing. These checks varied from once per day, to once per month, and longer than one month. The research during this period indicated that the calibration is stable over an extended period of time (> one month), but a definitive time could not be established. Due to this uncertainty, methods of forced self-testing were evaluated. ReporteR Reliability Testing Confirmed positive cocaine and methamphetamine samples previously analyzed in the first phase of laboratory testing were randomly selected and retested using the ReporteRs. Of 28 items of cocaine that were retested with a ReporteR device, 22 ReporteR results were consistent. Of 35 items of methamphetamine that were retested, 33 ReporteR results were consistent. Of the total samples tested, a consistency rate of ~87% was calculated. Consistent results could either be reported as negatives or positives. Testing was blind in that the randomly selected evidence was retested by various individuals, so samples may have been tested by the same individual and ReporteR device twice. Second Phase Methamphetamine and Cocaine Testing Prior to receiving the microscope, laboratory analysis was completed on field tested methamphetamine and cocaine evidence. ReporteR results for items suspected to contain methamphetamine and tested by officers had an accuracy rate of 77.9%. However, retesting by forensic scientists on those same items using the ReporteR yielded an accuracy rate of 100%. ReporteR results for items suspected to contain cocaine had an accuracy rate of 73.4% for both officer and lab testing. Page 37 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Correlations Testing Due to the low correlations found during cocaine testing, a study was completed by adjusting correlation values/number of bars using both the ReporteR alone and the NuSpec software. The results are outlined in Tables 6 and 7. Table 6. Correlation testing. (NuSpec software lists number correlations only; second version of library used; ‡inconclusive chemical field test) Sample Results Correlation/ # of Bars Device Alone or Software 5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.80, 2 bars = 0.79-0.70, 1 bar = 0.69-0.60, No match = 0.59-0.00 50/50 Cocaine/Caffeine Mix No Match ---ReporteR Cocaine 2 bars ReporteR Cocaine 0.7 NuSpec Cocaine/Sodium Bicarbonate 0.63 NuSpec Isopropylbenzylamine 0.52 NuSpec 60/40 Cocaine/Caffeine Mix No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR Cocaine 0.51 NuSpec Cocaine/Sodium Bicarbonate 0.47 NuSpec Pseudoephedrine 0.37 NuSpec 5 bars = 1.00-0.85, 4 bars = 0.84-0.75, 3 bars = 0.74-0.65, 2 bars = 0.64-0.60, 1 bar = 0.59-0.50, No match = 0.49-0.00 60/40 Cocaine/Caffeine Mix Cocaine 3 bars ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR Cocaine 0.35 NuSpec Cocaine/Sodium Bicarbonate 0.30 NuSpec Cocaine 1 bar ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR Cocaine 0.32 NuSpec Cocaine/Sodium Bicarbonate 0.28 NuSpec 5 bars = 1.00-0.80, 4 bars = 0.79-0.60, 3 bars = 0.59-0.50, 2 bars = 0.49-0.40, 1 bar = 0.39-0.30, No match = 0.29-0.00 60/40 Cocaine/Caffeine Mix Isopropylbenzylamine Caffeine Ephedrine Cocaine Cocaine Cocaine Cocaine No match Cocaine Cocaine Cocaine Cocaine No match Isopropylbenzylamine Isopropylbenzylamine Isopropylbenzylamine No match No match No match Ephedrine 2 bars 1 bar 1 bar 3 bars ---2 bars 3 bars 2 bars 2 bars ---5 bars 4 bars 5 bars ---------5 bars ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR Page 38 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Sample Results Correlation/ # of Bars Device Alone or Software 5 bars = 1.00-0.80, 4 bars = 0.79-0.60, 3 bars = 0.59-0.50, 2 bars = 0.49-0.40, 1 bar = 0.39-0.30, No match = 0.29-0.00 Ephedrine Ephedrine 5 bars ReporteR Ephedrine 5 bars ReporteR Ibuprofen Ibuprofen 5 bars ReporteR Ibuprofen 5 bars ReporteR Ibuprofen 5 bars ReporteR Lidocaine Lidocaine 4 bars ReporteR Lidocaine 4 bars ReporteR Lidocaine 4 bars ReporteR Mannitol No match ---ReporteR No match ---ReporteR No match ---ReporteR Lactose No match ---ReporteR No match ---ReporteR No match ---ReporteR Tetracaine Procaine 3 bars ReporteR No match ---ReporteR Procaine 3 bars ReporteR Diphenhydramine Calibration Standard 4 bars ReporteR Calibration Standard 4 bars ReporteR Calibration Standard 4 bars ReporteR Phentermine Ephedrine 4 bars ReporteR Ephedrine 4 bars ReporteR Lidocaine 1 bars ReporteR Procaine Procaine 5 bars ReporteR Procaine 5 bars ReporteR Procaine 5 bars ReporteR Ephedrine (different lot number) Ephedrine 5 bars ReporteR Ephedrine 5 bars ReporteR Ephedrine 5 bars ReporteR Ephedrine (different lot number) Ephedrine 5 bars ReporteR Ephedrine 5 bars ReporteR Ephedrine 5 bars ReporteR Amphetamine Amphetamine 5 bars ReporteR Amphetamine 5 bars ReporteR Amphetamine 5 bars ReporteR Levamisole Levamisole 5 bars ReporteR Levamisole 5 bars ReporteR Levamisole 4 bars ReporteR Benzocaine Cocaine 1 bar ReporteR Cocaine 1 bar ReporteR Cocaine 0.29 NuSpec 5 bars = 1.00-0.80, 4 bars = 0.79-.70, 3 bars = 0.69-0.60, 2 bars = 0.59-0.50, 1 bar = 0.49-0.40, No match = 0.39-0.00 Benzocaine No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR No Match ---ReporteR Benzoylecgonine‡ Cocaine 2 bars ReporteR Cocaine 2 bars ReporteR 5 bars = 1.00-0.90, 4 bars = 0.89-0.85, 3 bars = 0.84-0.80, 2 bars = 0.79-0.75, 1 bar = 0.74-0.70, No match = 0.69-0.00 Benzoylecgonine‡ Cocaine 1 bar ReporteR 5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.85, 2 bars = 0.84-0.80, 1 bar = 0.79-0.75, No match = 0.74-0.00 Benzoylecgonine‡ No Match ---No Match ---No Match ---No Match ---No Match ---No Match ---Cocaine 0.67 NuSpec software lists number correlations only; second version of library used; ‡inconclusive chemical field test ReporteR ReporteR ReporteR ReporteR ReporteR ReporteR NuSpec Page 39 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Sample Results Correlation/ # of Bars Device Alone or Software 5 bars = 1.00-0.95, 4 bars = 0.94-0.90, 3 bars = 0.89-0.85, 2 bars = 0.84-0.80, 1 bar = 0.79-0.75, No match = 0.74-0.00 Cocaine 0.64 NuSpec Cocaine 0.69 NuSpec Cocaine 0.70 NuSpec Cocaine 0.69 NuSpec Cocaine 0.71 NuSpec NuSpec software lists number correlations only; second version of library used; ‡inconclusive chemical field test Table 7. Field collected ReporteR correlations on conclusively identified cocaine street samples in the third phase of testing. Officer's Raman Result Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine (tested twice) Officer's Raman Result Cocaine Calibration standard Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Cocaine Officer Bars/Correlations 0.20 0.42 0.16 0.42 0.85 0.58 0.95 0.88 0.96 0.27 0.35 0.42 Officer Bars/Correlations 0.68 0.68 0.95 0.83 3 (>0.35) 1 (>0.35) 0.94 0.68 0.21 0.50 0.92 0.23 1 (>0.35) 0.58 0.74 0.41 0.36 GCMS Result Cocaine Cocaine Caffeine, cocaine Cocaine (sugars indicated) Cocaine (levamisole indicated) Cocaine Cocaine Cocaine Cocaine (levamisole indicated) Cocaine (levamisole, sorbitol, procaine indicated) Cocaine (levamisole, sorbitol, procaine indicated) GCMS Result Cocaine Cocaine (levamisole indicated) Cocaine Cocaine Cocaine (levamisole indicated) Cocaine (levamisole indicated) Cocaine, caffeine Cocaine Cocaine (phenacetin indicated) Cocaine Cocaine Cocaine (levamisole indicated) Cocaine (levamisole indicated) Cocaine (levamisole indicated) Cocaine (acetaminophen, lidocaine, levamisole indicated) Cocaine (levamisole indicated) Cocaine False Positive Field Testing Review A review of laboratory discovered field testing program errors revealed that from 2010 to the present, 71% of all chemical field tested cocaine errors encountered are due to false positives (lidocaine was conclusively identified in 60% of false positive results). See page 82 for LVMPD definitions and policy on field test errors. The spectra below (Figure 6) shows a ReporteR collected Page 40 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 lidocaine result with a correlation of 0.57. The white is the sample spectrum. The red is the lidocaine reference material spectrum. This sample was field tested positive for cocaine using a chemical field test kit. Figure 6. Lidocaine spectra. At the correlation settings used in the beginning of the research, a false positive for cocaine was discovered while testing a benzoylecgonine reference material. An electronic survey of Clandestine Laboratory Investigating Chemists (CLIC) members was completed regarding how often forensic laboratories have encountered benzoylecgonine without cocaine. Based on comments received from a world-wide, vast community of forensic chemists, it was found to be a rare occasion and was often found in association with decomposed/water-logged deceased persons or when the suspected cocaine had been wet for a substantial period of time. Figure 7 shows the spectra collected of benzoylecgonine with cocaine HCl and with cocaine base using the Raman microscope. The overlays are baseline corrected. The spectral similarities are apparent, however, each comparison has differences. Benzoylecgonine has two peaks at 1063 cm-1 and 1129 cm-1 that are not present in cocaine HCl. Cocaine base has the triplet at 848-898 cm-1 that is not present in benzoylecgonine. Page 41 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 7. Benzoylecgonine and cocaine spectra. Page 42 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 7. continued Page 43 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 7. continued ReporteR Self-test Field Testing During the second phase of testing, it was discovered that ReporteR field self-calibration checks would sometimes fail. In addition, detectives may not perform these calibration checks in the field. It was decided that calibration checks, or self-tests, would have to be controlled by the instrumentation to ensure proper operation. The ReporteRs were revised to include an automatic self-test notification that can be set by either length of time between self-tests or by the number of tests performed between each self-test. Routine testing of samples cannot continue until the check is completed. The notification is dictated by how many tests are performed per day by a detective. To ensure that the instrument is checked with an average frequency of at least once per day, the automatic self-test notification is currently set to prompt for a calibration check to occur once every three tests that are performed. Microscope Validation The microscope’s spectral resolution was able to show finer, more detailed differences between spectra, especially when mixtures were tested. None of the negative controls analyzed Page 44 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 matched to cocaine or methamphetamine. The positive controls demonstrated the microscope’s ability to detect methamphetamine and cocaine in street samples. Only one sample of street cocaine had a library match to benzoylecgonine, although when viewing the spectrum, it was apparent that the sample actually contained cocaine. Third Phase Methamphetamine and Cocaine Testing Since installation of the Raman microscope, a total of 130 methamphetamine cases were tested in the field, the majority of which were completed using the enhanced ReporteR. Out of the total of 130 methamphetamine cases tested, four had color field test errors or a color field test was not completed, two did not include a ReporteR print out in the case, several test results could not be used due to equipment malfunctions (see Errors, Malfunctions, and Repairs below), and 75 were field tested with the repaired enhanced ReporteR and analyzed by laboratory personnel. Of these 75 cases, 73 case samples field tested presumptive positive for methamphetamine using the repaired enhanced ReporteR. One sample tested positive for amphetamine (0.59 correlation) and one tested positive for lidocaine (0.35 correlation). Both were confirmed to contain methamphetamine with no indications of amphetamine or lidocaine. The methamphetamine field testing accuracy rate with an enhanced ReporteR is 97.3% for the 75 cases. The methamphetamine laboratory accuracy rate was 96% with a non-enhanced ReporteR. To date, a total of 27 cocaine cases have been tested with an enhanced ReporteR field testing accuracy rate of 100%. The laboratory accuracy rate was 66% with a non-enhanced ReporteR. It does not appear that the field ReporteR equipment malfunctions affected the device’s ability to test cocaine; however, fewer tests were performed on suspected cocaine samples. Errors, Malfunctions, and Repairs Five cocaine cases resulted in positive ReporteR field results; however, 4 of these cases had correlations of 0.42 or less (out of 1.00). Thirty-one of 33 methamphetamine cases resulted in positive ReporteR field results using the enhanced ReporteR. Five of the 31 cases resulted in at least one library mismatch. Four of these mismatched cases gave the officer a result of cocaine with a correlation of 0.26 or less. After each cocaine mismatch, the officer repeated the test and received a final result for methamphetamine with a correlation of 0.40 or above (each item is not to be tested more than three times). One sample gave the officer a mismatch for lidocaine twice, and a final result of pseudoephedrine with a correlation of 0.21. These types of errors had not been encountered in previous testing when using the non-enhanced ReporteRs. It was determined by the manufacturer Page 45 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 that these errors were due to equipment malfunctions. Simple repairs were applied by the laboratory and the firmware was updated by DeltaNu, Inc. (SciAps, Inc.). Methamphetamine Spectral Comparison The following spectra in Figure 8 were collected on methamphetamine street samples. In each case, the first spectrum was collected by a detective in the field (F) using a ReporteR. The second (and third in one case) was collected in the laboratory (L) using a ReporteR. The last spectrum in each case was collected using the Senterra Raman microscope. The correlations reported by the ReporteR devices are listed below each spectrum. Figure 8. Spectra of methamphetamine samples. Case #1 (F) Correlation 0.44 methamphetamine Page 46 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #1 continued (L) Correlation 0.84 methamphetamine Sample on top in red Reference material on bottom in blue Page 47 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Spectra of methamphetamine samples (continued). Case #2 (F) Correlation 0.83 methamphetamine (L) Correlation 0.84 methamphetamine Page 48 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #2 continued Sample on bottom in red Reference material on top in blue Page 49 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Spectra of methamphetamine samples (continued). Case #3 (F) Correlation 0.87 methamphetamine (L) Correlation 0.84 methamphetamine Page 50 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #3 continued Sample on bottom in red Reference material on top in blue Figure 8. Spectra of methamphetamine samples (continued). Case #4 (F) Correlation 0.36 methamphetamine Page 51 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #4 continued (L) Correlation 0.38 lidocaine (L) Correlation 0.88 methamphetamine Page 52 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #4 continued Sample on bottom in red Reference material on top in blue Figure 8. Spectra of methamphetamine samples (continued). Case #5 (F) Correlation 0.83 methamphetamine Page 53 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 8. Case #5 continued (L) Correlation 0.72 methamphetamine Sample on bottom in red Reference material on top in blue Cocaine Spectral Comparison The following spectra windows (labeled 1 through 8) in Figure 9 were collected on cocaine street samples. The spectra with black backgrounds were collected in the field with correlations labeled underneath each numbered spectrum. In windows 1 and 2, pre- and post-baseline corrected spectra are shown. Windows 3 and 4, windows 5 and 6, and windows 7 and 8 each Page 54 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 show spectra from a set of samples collected by the ReporteR and then by the microscope. In window 8, cocaine spectra collected on the same substance in two types of packaging is shown. The same sample tested through gray, opaque colored plastic yielded a laboratory ReporteR correlation of 0.85 (spectrum not shown). Figure 9. Spectra of cocaine samples. 1. Before baseline correction Sample on bottom in red Reference material on top in blue 2. After baseline correction Sample on bottom in red Reference material on top in blue Page 55 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 9. Cocaine spectra continued 3. Field Correlation 0.58 4. Sample on bottom in red Reference material on top in blue Page 56 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 9. Cocaine spectra continued 5. Field Correlation 0.21 Sample in middle in red Cocaine reference material on top in blue Alpha Lactose reference material on bottom in black 6. Page 57 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 9. Cocaine spectra continued 7. Field correlation 0.36 Cocaine clear plastic Gray plastic on top in black Clear plastic on bottom in red 8. Microscope of clear plastic versus gray plastic Page 58 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Cocaine and Diluent Comparison A comparison of cocaine to common diluents was completed using the Raman microscope. The compounds that were compared in Figure 10 (caffeine, lactose, levamisole, lidocaine, and mannitol) had several peaks that may cause interference. Two peaks (1715 cm-1 and 1025 cm-1) appear to be unique to cocaine versus levamisole, mannitol, lidocaine, caffeine, and lactose. Although the peak at 1025 cm-1 is unique to cocaine as compared to these diluents, it is not as strong as others in the cocaine spectrum. There also may be possible interference with both the 1715 cm-1 and the 1025 cm-1 from close adjacent peaks when the cocaine is mixed with caffeine, levamisole, or mannitol. Caffeine has a strong Raman signature, and the interference of this substance with cocaine is shown in a baseline corrected direct overlay below. Figure 10 is shown on the next page. Page 59 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 10. Cocaine and diluents spectra. Page 60 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 11 below demonstrates the importance of testing a sample multiple times. Lab ReporteR testing indicated that acetaminophen, lidocaine, and cocaine were present in the sample when tested three consecutive times on three different spots (first lab ReporteR result: 0.82 acetaminophen, second lab ReporteR result: 0.62 lidocaine, third lab ReporteR result: 0.33 cocaine). The first spectrum was collected by the detective in the field. The Raman microscope was able to detect both acetaminophen and cocaine which is indicated in the spectra below. Acetaminophen, lidocaine, some form of amisole, and cocaine were confirmed to be contained in the sample via GC/MS. Figure 11. Spectra of cocaine mixed with acetaminophen, lidocaine, and levamisole. Field Correlation: 0.41 Cocaine Page 61 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 11. continued Sample in middle in black Cocaine reference material on top in red Acetaminophen reference material on bottom in blue ReporteR Marijuana Testing No viable spectra (spectra with peaks or features that could be differentiated) were obtained during plant material and tetrahydrocannabinol testing with the ReporteR. No demonstrative peaks were apparent in the spectrum compared to a blank run. The spectrum of the delta-9tetrahydrocannabinol/extracted filter/water exhibited fluorescence as compared to a water blank. Testing of the marijuana secondary reference material and the secondary reference plant materials (eucalyptus, catnip, “hydro”, tobacco, psilocin mushrooms, and culinary mushrooms) demonstrated similar spectra, revealing a large amount of interfering fluorescence. After the aforementioned solvents and solutions were tested for Raman response, water, 1.0 N NaOH, 2.0 N HCl, and methylene chloride were found to yield the Raman spectra with the least interference (no peaks or very few). As most cannabinoids cannot be extracted from marijuana with water alone, the other three agents were used to solvate a secondary marijuana reference material (the solvent was filtered into liquid sample vials). Spectral differences were evident in the blank versus marijuana extractions, however the marijuana extraction spectra appeared much like the Page 62 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 directly tested marijuana plant material with few distinguishable characteristics. TLC testing revealed similar results. The spectrum of the marijuana sample, taken at the delta-9tetrahydrocannabinol Rf, looked much like that of the TLC plate blank spectrum. Viable spectra of the delta-9-tetrahydrocannabinol reference material on an aluminum slide were not obtained using the ReporteR. Microscope Marijuana Testing The 532 nm laser did not produce viable spectra and its use was discontinued during the testing. As has been dictated by Raman spectroscopy at this wavelength, glass caused fluorescence interference with the 785nm laser. Metallic based slides provided the best spectra. Spectra were obtained at 15 second integration with 3 co-additions, 10 seconds with 2 co-additions, and 5 seconds with 2 co-additions. Each provided sufficient information, so the remainder of testing was carried out using 5 second integration with 2 co-additions, as this was the minimum that was sufficient for testing and resembled the collections taken by the ReporteR. A glass slide covered in aluminum foil provided a cost effective and quick solution as a base for cannabinoid and marijuana testing. Testing random areas of plant material using the Raman microscope was not reliable in terms of detecting cannabinoids. In most instances, the fluorescence was overpowering. Several attempts were made on many of the samples to determine if there were actual microscopic features of the plant material that yielded viable spectra. Such features were identified. These features are best described as globular, amber colored resinous structures that can be smooth or striated in appearance (Figure 12). Structures that appeared more transparent in color, crystalline, or less smooth did not produce viable spectra (Figure 13). Page 63 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 12. Viable microscopic features of marijuana. Figure 13. Non-viable microscopic feature and spectrum of marijuana. Page 64 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Eight of the ten case work plant materials tested yielded spectra that contained features in common with cannabinoids, specifically cannabinol and delta-9-tetrahydrocannabinol. When tested using GC/MS, delta-9-tetrahydrocannabinol and cannabinol were predominant in the samples. One sample tested indicated a strong presence of cannabinol in the Raman spectrum (Figure 14). This spot of the slide was isolated, extracted with methanol, and analyzed with GC/MS. The results yielded a higher abundance of cannabinol according to the TIC (not quantitated), than was previously seen in an earlier GC/MS test of a random sample selection of the same plant material (Figure 15). Figure 16 is a photo of the feature tested. Figure 14. Marijuana and cannabinoid spectra (continued on next page). The first window shows the sample spectrum prior to baseline correction. The second window shows the sample spectrum at the top, cannabinol reference material in the middle, and delta-9tetrahydrocannabinol on the bottom. Both cannabinol and delta-9-tetrahydrocannabinol are indicated in the sample, however, the strong influence of cannabinol is apparent. Spectra in the second window were baseline corrected. Page 65 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 14. continued Sample before baseline correction Sample (post-baseline correction) on top in black Cannabinol reference material in middle in blue THC reference material on bottom in green Page 66 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 15. Cannabinoid total ion chromatograms. Figure 16. Microscopic feature. Out of the negative plant materials investigated, lantana, honeysuckle, comfrey, horehound, mugwort (Figure 17), thistle, and oregano had microscopic features that were similar in appearance to marijuana crystalline or resinous structures. Horehound and oregano (Figure 18) both yielded spectra that were dissimilar to the cannabinoid spectra that had been obtained previously using the Raman microscope. Page 67 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 17. Mugwort photo and spectrum. Page 68 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 18. Oregano. Photo, pre-baseline corrected spectrum, and post-baseline corrected spectra as compared to delta-9-tetrahydrocannabinol. Oregano before baseline correction Oregano baseline corrected on top in green THC on bottom in black Page 69 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Black Tar Heroin Testing Black tar heroin was also investigated using various methods that may be translated to a field testing program. The LVMPD Forensic Laboratory was introduced to Thermo Scientific’s portable Raman device, the TruNarc. Testing black tar heroin using the TruNarc requires a sampling procedure that begins with dissolving the sample in ethanol and evaporating the solution onto a metallic wafer sampling device. Using the ReporteR on black tar heroin with Thermo’s heroin sampling system (metallic wafer and ethanol) yielded similar spectra to that collected using the TruNarc. These spectra appear to lack specificity as compared to spectra of a traceable heroin reference material obtained by the Bruker Senterra Raman microscope. Black tar heroin was tested using various techniques, including using graphene based substrates to quench the fluorescence of the dark black-brown substance. The graphene testing did not produce strongly defined spectra. However, ReporteR tested samples showed possible indications of peaks. Figure 19 illustrates the ReporteR collected spectra (reviewed using Bruker OPUS software) compared to a blank graphene/glass fiber substrate (second window), and compared to a spectrum collected using the Raman microscope (third window). The first window shows the heroin reference material collected by the microscope. The microscope did not produce a viable sample spectrum, regardless of preparation. Use of the baseline correction technology did not aid much in the reduction of the background, likely due to its extremely dark color and other compounds present in the naturally-derived, semi-synthetic substance. Other solvation/evaporation and potassium bromide mixing experiments did not produce viable spectra. Page 70 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 19. Black tar heroin spectra. Heroin reference material (microscope) Blank on bottom in black (ReporteR) Heroin/graphene samples in blue and green (ReporteR) Page 71 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 19. continued 2 ReporteR heroin/graphene samples (blue and green) overlaying microscope collected heroin/graphene sample (dark red) Ecstasy Testing Limited testing on ecstasy tablets was also completed. The testing of such items reveals that spectra of the diluents present in the tablets are most likely to be collected, however if focused on a crystal within the tablet that contains methylenedioxymethamphetamine, it can be individualized by the Raman microscope, even on colored tablets. Figure 20 shows a prebaseline corrected spectrum, and a comparison of the corrected spectrum to a methylenedioxymethamphetamine reference material spectrum. This sample was a light, mint green tablet in a plastic bag. Spectra were collected using the Raman microscope for case samples consisting of a yellow-green tablet, a yellow tablet, a red tablet, a blue tablet, a darker mint green mottled tablet, a bright green tablet, and a clear capsule containing brown granules. The samples contained various substances that were confirmed via GC/MS. Although spectra exhibited fluorescence and interference from the tablet mixtures, all could be baseline corrected and most substances found during conclusive analysis could be identified from the spectra collected. ReporteR testing was completed on two of the above samples and no viable spectra were collected (only a high baseline). Page 72 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 20. Ecstasy/MDMA spectra. Sample pre-baseline correction Baseline corrected sample on bottom in blue MDMA reference material on top in black Page 73 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Synthetic Drug Testing A few reference materials of designer synthetic drug analogs were evaluated. The compounds did produce spectra and it was indicated that the amount of fluorescence or other background interference was dependent on isomeric form. The baseline correction technology proved useful in deconvoluting these spectra (Figure 21). This information was shared with the Kentucky State Police Central Forensic Laboratory in regards to the research they have completed. UR-144, XLR-11, AM-2201, methylone, 4-methylethcathinone, methylenedioxypyrovalerone, pyrrolidinopentiophenone, butylone, 2- and 3-fluoroamphetamine, 4-fluoromethcathinone, chlorophenylpiperazine, fluorophenylpiperazine, and 2-,3-, and 4-fluoromethamphetamine all produced spectra using the Raman microscope, with or without baseline correction. Only UR-144, XLR-11, methylone, 4-methylethcathinone, methylenedioxypyrovalerone, pyrrolidinopentiophenone, butylone, and pyrrolidinopentiophenone were tested using the ReporteR. 4-methylethcathinone exhibited some interference from fluorescence but peaks were detectable. Pyrrolidinopentiophenone (alpha-PVP) exhibited such a high baseline that it could not be differentiated using the ReporteR device alone. Figure 21 is shown on the next page. Page 74 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 21. Synthetic drug spectra. 3- fluoroamphetamine pre-baseline corrected on top 2- fluoroamphetamine on bottom 2- fluoroamphetamine on top 3- fluoroamphetamine post-baseline corrected on bottom Page 75 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 21. continued Pre-baseline correction Alpha-PVP ReporteR on top, alpha-PVP microscope on bottom Post-baseline correction Alpha-PVP microscope on top in black, alpha-PVP ReporteR on bottom in red Page 76 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Figure 21. continued Pre-baseline correction AM-2201 powder on bottom in blue AM-2201 plant material on top in red Post-baseline correction AM-2201 powder in blue AM-2201 plant material in red Page 77 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 IV. Conclusions 1. Discussion of Findings ReporteR Enhancements Enhancements of the ReporteR which included suggestions made by LVMPD to the manufacturer were well received by the detectives using the equipment in the field. The display screen was changed to no longer show a bar graph representing the correlation. A number correlation is listed out of 1.0. The implementation of a portable Bluetooth printer and printout made the detective’s job easier and simplified the testing procedure. Incorporation of a forced self-test that required the detectives to perform self-tests more often not only alleviates issues of forgetting to perform the calibration check, but also informs them of possible immediate malfunctions, and prevents them from submitting test results after a failed self-test. The self-test results are also included on the printout, supplying the court with proof that a recent calibration check was performed. Changing the design of the polystyrene calibration standard also simplified the process, as detectives no longer have to switch attachments in order to perform the testing. There were no complaints of the attachment being loose on the device after the magnet was replaced. ReporteR Results Display Screen The forensic scientists performing the testing had the opinion that only showing “Methamphetamine” “Cocaine” and “No Match” results with a bar correlation provided the most effective way of communicating results to an officer. However, detectives preferred the screen display that reports out what the substance might be, even if the results cannot be used for preliminary hearing. For example, the screen may read out “MDMA,” and even though only methamphetamine and cocaine results may be used in court, the detectives felt that the information would aid them in their investigation. The correlation display, bars versus numbers, was inconsequential as long as it was understood that 1.00 was the highest correlation possible. However, it is the opinion of the laboratory personnel that a bar graph representation would prevent erroneous results being submitted to court, as these values can be regulated. Self-testing During laboratory testing, calibrations appeared to be stable for over a month. However, once the devices were placed in the field and checked periodically by lab personnel, self-tests began to fail at varying lengths of time. This may be due to the conditions that the devices were being Page 78 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 subjected to (extreme heat, being handled by detectives, etc.). It is difficult to perform laboratorybased calibration checks on all the ReporteR devices being used by officers for field testing on a strict schedule due to timing, location, and availability. To ensure the calibration checks are routinely performed, the ReporteRs include a forced self-test that is pre-set by laboratory personnel. The officers were trained on how to perform the check using the polystyrene standard. Proof of the checks is included on the results printout. ReporteR Sampling Attachment Design Two different ReporteR sampling attachment designs were tested in the field. The first design incorporated an adjustable focus that could be changed depending on the thickness of packaging. Since the packaging often consisted of single polyethylene plastic bags, an attachment was designed by DeltaNu, Inc. (SciAps, Inc.) that was not adjustable and had a stable focus for this type of packaging. As a higher number of mismatches and “no signal” results were received with this design, the ReporteR was reverted to incorporate the original adjustable attachment. Officers did not like the chain that was attached to prevent loss. It was requested that the manufacturer incorporate a better method of holding the attachment to the device, and subsequently, they increased the strength of the magnet that holds the attachment. Libraries Prior to the National Institute of Justice grant award, reference materials were tested against the large library included with the ReporteR device. Testing also included laboratory prepared libraries. It was determined that a lab-created library of known reference materials would be the most prudent for a field testing program situation where only a few selected substances were being reported. To present instrument analysis results in an evidentiary hearing, a known set of traceable reference materials entered by the scientists performing the research and testifying on the findings was determined to be a more appropriate reference than using a manufacturer’s library. It was also noted that including mixtures of reference materials caused false results during testing. After receipt of the grant award, it was established that some secondary reference materials loaded into the library (confirmed street samples) caused false results during testing. It was also found that the procaine reference material in the library was causing mismatches for methamphetamine, and was subsequently removed from the library. As procaine is not as often encountered in evidence submissions in southern Nevada, it was not an issue. Caffeine, which is Page 79 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 commonly seen in cocaine samples, has a strong Raman signal. Therefore, it was not included in the final version of the ReporteR library as cocaine was less likely to be detected when mixed with caffeine. It is a recommendation that any Raman-based field testing program, that will need to be approved in court prior to use, establish their own library of reference materials to better suit the needs of the region and the drugs commonly encountered in the area. Repeatability Although the consistency rate was expected to be higher than 87%, it is understood that an individual will not be able to test the exact same crystals tested previously. This affects the likelihood of producing the exact same results, especially when street samples tend to be mixtures of a drug and a diluent. False Positives For the purposes of a Raman-based presumptive field testing program, a false positive is defined as any sample yielding a positive result for the target substance (i.e. for what is being tested for in the program), when that sample does not contain that substance. Any acceptable correlation to a substance other than cocaine and methamphetamine (or any target compound) would result in the evidence being submitted to the laboratory for analysis, much like in the current chemical field testing program, in which interpreting color is subjective and not always accurate. Although one false positive (benzoylecgonine) for cocaine had been discovered during laboratory reference material testing, the observance of this substance in casework testing is rare and encountered only in unique circumstances. (False positive results for benzocaine were also discovered, however, this was during the correlation testing when the correlations were purposely set very low). A few false positive mismatches occurred during the third phase of methamphetamine testing (methamphetamine samples resulting in low correlation cocaine matches), but these were determined to be due to equipment malfunctions and did not occur after the equipment was repaired. A preventative measure to avoid having these results submitted to preliminary hearings would be to set the correlation level such that a result under a certain correlation could not be used for court purposes, and to require that suspected cocaine samples that were discovered under wet or degraded conditions be submitted directly for laboratory analysis. Equipment Malfunctions During the initial evaluation of the enhanced ReporteR in the field, result mismatches complicated the research. Upon evaluation of the spectra, it was evident that there may have been Page 80 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 instrument malfunctions. These data files were sent to DeltaNu, Inc. (SciAps, Inc.) for review and it was determined that the issues were due to possible light interference and focusing issues. These problems were addressed by performing routine maintenance on the ReporteR and reverting back to an earlier version of the sampling attachment. These false results may be mitigated by requiring that positive results have correlations over a set number, or by requiring that a non-methamphetamine or a non-cocaine result be submitted to the lab for confirmatory testing. Methamphetamine In the first phase of laboratory methamphetamine testing, the ReporteR accuracy rate was 92%. Chemical color field testing of suspected methamphetamine evidence in the lab and in the field resulted in an accuracy rate of 100%. It was determined, after receiving these results, and finding that the prevalence of isopropylbenzylamine had waned, that methamphetamine chemical field test kits would no longer be performed in the remaining laboratory portion of the testing (but would continue in the field) unless a false positive field result was discovered. The second phase of methamphetamine field testing revealed an accuracy rate of 77.9%. The lab accuracy rate was 100%. In the third phase of testing, suspected methamphetamine cases were tested in the field using an enhanced ReporteR with an accuracy rate of 100%. The laboratory accuracy rate for testing the same samples with a non-enhanced ReporteR was 96% during this phase. It is concluded from this trend that the enhanced, upgraded device (after repairs) caused an increase in accuracy for officer field testing for methamphetamine, which may be partly due to ease of use, but is most likely due to an increase in the signal-to-noise ratio. This trend is also evident in the cocaine data. Upon reviewing spectra of methamphetamine collected in the field using the ReporteR, and in the lab using a ReporteR and a microscope, there was no notable trend in field data versus lab data collected. Microscope collected spectra were obviously more detailed. ReporteR collected spectra were mostly adequate, but varied in quality. In fact, of the spectra examples shown in Figure 8, the best spectrum was collected in the field. It is evident that spectra collected may vary based on sampling technique and the quality of signal received by the ReporteR. The ReporteR also shows higher baselines in its spectra compared to the microscope. Lidocaine mismatches were due to the library spectrum of lidocaine as seen in Figure 8. Any spectrum collected with a similar baseline may result in a match to lidocaine, which could explain the incidence of lidocaine mismatches for Page 81 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 methamphetamine samples. An automatic baseline correction method added to an enhanced portable Raman device would be beneficial against this hindrance. Cocaine A historical review of cocaine chemical field tests performed by officers in the field revealed flaws in the implementation of testing and/or interpretation of results. LVMPD’s current chemical field testing program tries to ensure quality by requiring that apparent errors made by officers during testing are recorded, the officers and chain of command are notified, and remedial training may be required. Across the board, most errors are clerical, as is indicated on the checklist that officers are required to complete. Some errors are defined as false positives. Even though the chemical field tests for cocaine are formulated such that very few substances should give a false positive result, if the test is performed incorrectly (overloading) or interpreted incorrectly (blue color appears in the wrong step), an erroneous positive result may be recorded. In such cases where a false positive is suspected or no cocaine is identified in a sample, an identical field test is performed by laboratory personnel. If the results cannot be replicated in the lab, this is considered an error. All errors that are noticed by the Clark County, Nevada District Attorney’s Office are sent to the lab as a confirmatory analysis is needed prior to trial. As the Forensic Laboratory only sees a small portion of these field results, the true percentage of errors due to false positives is unknown. Use of a portable Raman field testing program would virtually eliminate such errors, including subjective color interpretation which can lead to false positives, and the checklists would be replaced by a printout reporting the results. The lower the correlations are set, the more often a positive result is obtained. However, the possibility of receiving false positives increases. When the lowest number of bars was set to 0.30, a benzocaine reference material tested positive for cocaine using the ReporteR. Benzoylecgonine had correlation values as high as 0.71, testing positive for cocaine. The comparison of spectra collected using the Senterra showed the similarities between cocaine HCl, cocaine base and benzoylecgonine, but differences were notable. Cocaine samples mixed with various diluents also often caused results with low correlation values. According to the results shown in Table 7, if the minimum correlation value was set above 0.71 (in order to avoid false positive results with benzoylecgonine), ~74% of accurate positive cocaine results from the ReporteR would be eliminated. However, if wet or suspected degraded cocaine samples were not tested during the field testing (to avoid possible Page 82 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 benzoylecgonine samples), and the correlation levels were set above 0.35 (to avoid benzocaine false positives), only ~15% of results would be eliminated from being used for court purposes. As shown in Figure 9, a cocaine sample tested using the Senterra matched to diphenhydramine after performing a library search. However, after baseline correction, the sample then matched to cocaine. This sample was found to contain levamisole and cocaine hydrochloride using GC/MS and FTIR/ATR (Fourier transform infrared/attenuated total reflectance). The baseline correction algorithm utilized by the Raman microscope aided in detection of cocaine for this sample. The ReporteR does seem to have some advantages when it comes to testing samples in which the pinpoint accuracy of the microscope would be a detriment. In Figure 9, cocaine spectra collected on the same substance in two types of packaging is shown. The sample tested through gray, opaque colored plastic yielded a lab ReporteR correlation of 0.85. The spectrum collected on the microscope as shown in the eighth window of the figure, shows the high baseline generated from testing the gray plastic packaging, as the Senterra’s laser caused sample burning. As with methamphetamine, the quality of ReporteR spectra varied (Figure 9, windows in the 3, 5, and 7 positions). The results of cocaine testing are far less accurate than the methamphetamine testing. This is likely due to the spot size and amount and assortment of diluents used to cut cocaine. When comparing cocaine to common diluents encountered in the laboratory, there are peaks that are unique to cocaine and may assist in creating a method to increase detection of cocaine in mixtures. Caffeine, in particular, may cause a great amount of interference, especially around the two abundant cocaine peaks at 1598 cm-1 and 1715 cm-1. Caffeine also has two peaks in this area at 1600 cm-1 and 1699 cm-1. Development of methods to isolate peaks specific to cocaine as compared to caffeine may be necessary. In the first phase of laboratory cocaine testing, the ReporteR accuracy rate was 73%. Chemical color field testing of suspected cocaine evidence in the lab and in the field also resulted in an accuracy rate of 73%. Due to these results, it was determined that cocaine chemical field tests would continue to be performed in the remaining laboratory portion of the testing. In the second phase of testing, ReporteR results for items suspected to contain cocaine had an accuracy rate of 73.4% for both officer and lab testing. In the third phase of testing, a total of 27 suspected cocaine cases were tested in the field using an enhanced ReporteR with an accuracy rate of 100%. The laboratory accuracy rate for testing the same samples with a non-enhanced ReporteR was 66% during this phase. As indicated from this trend, the enhancements made to the ReporteR Page 83 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 increased testing accuracy for cocaine. It does not appear that the field ReporteR equipment malfunctions affected the device’s ability to test cocaine; however, fewer tests were performed on suspected cocaine samples. It is decidedly unknown why cocaine did not seem to be as affected by the malfunctions, but it may be due more to the nature of the methamphetamine data and it’s interaction with the programming and library matching algorithms. Although the officer detected cocaine at a 0.41 correlation, lab ReporteR testing indicated that acetaminophen, lidocaine, and cocaine were present in a sample when tested three consecutive times on different spots (first lab ReporteR result: 0.82 acetaminophen, second lab ReporteR result: 0.62 lidocaine, third lab ReporteR result: 0.33 cocaine) (Figure 9). The microscope testing was able to detect both acetaminophen and cocaine. Acetaminophen, lidocaine, levamisole, and cocaine were confirmed to be contained in the sample via GC/MS. This supports that a sample may need to be tested multiple times for an accurate positive result, however, three times was determined to be the limit in order to simplify testing procedures and decrease the possibility of errors. Marijuana Repeatable spectra from marijuana plant material have been collected using the Bruker Senterra Raman microscope and spectral attributes in common with cannabinoids were evident, specifically cannabinol and delta-9-tetrahydrocannabinol. Two samples of marijuana, of the ten tested, did not produce any viable spectra from the features found. At this stage, portable Raman field testing of marijuana is not practical due to the necessity of the microscope to locate the resinous structures in which the cannabinoids were detected in situ. However, the use of the baseline correction function of the Raman microscope’s software proved effective in reducing the fluorescence background in the spectra and would prove useful if automatically applied in a portable device. Testing using the ReporteR device included various methods of trying to extract cannabinoids from plant material in preparation for Raman testing. This negates the purpose of trying to stray from the use of hazardous chemicals, and this research was discontinued. Heroin Although a few weak repeatable spectra were collected using the ReporteR and graphene, it is evident that more research is needed in order to directly use portable Raman technology for the presumptive analysis of black tar heroin. Indications of a peak at 625 cm-1 and the triplet at 869908 cm-1 are seen in the ReporteR spectra as compared to the reference material analyzed using the microscope. It is the opinion of the researcher, that spectra collected using the TruNarc Page 84 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 sampling system are identifiable, but less than ideal, and the future of using Raman spectroscopy to analyze black tar heroin seems dependent on using some sort of liquid sampling system. Ecstasy Contrary to marijuana and black tar heroin, testing ecstasy tablets directly is possible with Raman-based portable technology, but limited. Ecstasy tablets tend to come in all colors and are made with various binding materials, both of which cause fluorescence issues. They also can contain various compounds, not only methylenedioxymethamphetamine, which is the historically typical substance encountered in ecstasy. The concentration of drug can vary widely, and the recent emergence of new synthetic compounds has complicated the issue further, as many of these compounds have been detected in supposed ecstasy tablets. By testing different colors of samples containing various components, it can be concluded that the baseline correction technology is useful in the reduction of fluorescence caused by the colorings and bindings present in the tablets, and aids in identification of spectra. Synthetics Although it is unlikely that a field testing program will be useful in identifying the emerging synthetic drug analogs (i.e. cathinones, phenethylamines, cannabimimetics) due to their strong molecular similarities, superfluity, and constant structural manipulation, indications that these drugs may be present can be determined by portable Raman spectroscopy. Testing a few selected reference materials with the Raman microscope determined that isomeric form may be differentiated using this type of instrumentation, and fluorescence interference may be based on the isomer. A few reference materials exhibited high baselines when tested with the ReporteR, but these spectra could be corrected using baseline correction technology. 2. Implications for Policy and Practice Enhancing and implementing a new method of presumptive field testing will benefit police officers, law enforcement field workers, forensic laboratories, and the court system. These benefits include safety, cost saving, time saving, and expedition of the judiciary process. A new method of field testing will enhance the ability of the law enforcement community to identify, analyze and interpret controlled substance evidence using a robust tool that gives immediate and defendable preliminary results for cocaine and methamphetamine, and other drugs if possible. Using a portable Raman device will take the subjectivity out of field testing, which will have an impact on exonerations. By completing this research, the laboratory ultimately hopes to Page 85 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 establish case law that may assist other local and state agencies to utilize this technology in the field. This scientific evaluation and justification for the court system is of utmost importance. Because most police officers do not have the expertise to evaluate spectra and to explain to the trier of fact how the technology works, a handheld device such as the ReporteR can give officers the results they need to present in court. In order for this to happen, the court must first be presented with scientific proof to satisfy the Frye standard that a Raman handheld device is a valid and reliable tool for preliminary identification of controlled substances. This has other implications for judiciary proceedings. Contamination is a common question from defense attorneys when testifying as to the validity of the field color tests. This research has also shown that there are discrepancies between officer and forensic scientist interpretations of chemical field test results. Not only would contamination be less of an issue with the Raman testing procedure, the incidence of the false positive results that occur with chemical field test misinterpretations might be eliminated, and the officers’ results in lieu of analysis may be used in preliminary hearings, saving time for both the laboratory and the court. The following illustrates an example of the difficulties experienced by officers in the field pertaining to the interpretation of colors observed when performing field tests. A submission which had field tested positive for cocaine by an officer was tested as part of our research. The chemical field test and ReporteR tests performed in the laboratory both gave negative results and the GC/MS data concluded the substance contained only lidocaine with no cocaine detected. This discrepancy may be attributed to an officer overloading a sample that contained only lidocaine and inattention to intermediate color changes in the chemical-based field test which led to an incorrect positive result for cocaine. As indicated by the research, cocaine testing using portable Raman technology may be somewhat limited. The correlation settings on the ReporteR may be set in order to prevent false positive occurrence, but this also hinders the ReporteR’s ability to accurately detect the target compound in mixtures. This may slightly increase the number of suspected cocaine submissions to the laboratory, due to more inconclusive results, but it significantly decreases the chance of false positive results. Law enforcement agencies spend tens of thousands of dollars on color test field kits every year. By switching to a portable Raman-based method for testing cocaine and methamphetamine, over $30,000 could be saved per year (after initial costs). Cost is a major consideration for most policing agencies, especially in the current economic climate. In addition, the training for this Page 86 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 technology could be cost effective and easily obtainable by providing free online video instruction. By testing the handheld devices in the lab, in the field, and by utilizing a Raman microscope, the LVMPD Forensic Laboratory has determined that portable Raman technology is an effective and reliable tool to presumptively identify methamphetamine and cocaine in the field. The laboratory has also identified which characteristics of portable Raman technology need enhancement in order to be implemented in a presumptive field testing program. The field testing program will incorporate the testing procedures that were found to be effective during the research. The laboratory would like to compile the data and present in court before the end of 2014. During the 2014 year, the enhanced ReporteRs currently in place at the LVMPD will be used by the narcotics detectives. Also, during this time, a training program will be developed that adheres to LVMPD policy and will include easily accessible web-based training. The training program will address the strengths and weaknesses of Raman technology so that officers fully understand the limitations and advantages of the testing. Future incorporation of more devices to be placed at substations to be used by patrol officers will commence after the technology has been accepted by the court system and the training program has been finalized. The LVMPD will provide the training program and procedures for other jurisdictions to model. If other jurisdictions wish to implement this type of presumptive testing program, the researchers suggest that a definitive testing procedure be laid out for officers to follow and that proof of results (some sort of printout) can be provided in court. There are four important aspects that should be incorporated in a Raman-based testing program. (1) A minimum/maximum number of tests per sample, (2) A printer or a device with printing capabilities, (3) Limiting testing to suspected substances in which literature and/or manufacturers can provide proof of repeatable, accurate results, and (4) The local courts should understand and accept the accuracy and limitations of the technology. 3. Implications for Further Research The use of the baseline correction technology on the Raman microscope was essential in the collection of spectra of compounds that inherently produce fluorescence, namely ecstasy, marijuana, and synthetics. A high baseline was also evident in methamphetamine and cocaine spectra when tested using the ReporteR. Baseline correction technology is available with software applications, however, this requires that a Raman device be coupled to a computer. Research needs to be Page 87 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 completed by those with the abilities to incorporate an automatic baseline correction algorithm into field based technology. Baseline correction was indispensable when detecting cannabinoids in marijuana. Although the current prospect of being able to use non-microscope based Raman technology on marijuana is unconvincing, the idea of detecting tetrahydrocannabinol and cannabinol in situ using spectroscopy opens the door for other methods of confirmatory forensic drug testing. Further research on marijuana will be completed by this laboratory using Raman spectroscopy. Automatic baseline correction may not prove valuable against caffeine, but it may help in deconvoluting spectra containing other contaminants. An algorithm based method may be created to assist in Raman analysis of cocaine mixtures. The analysis of mixtures may also benefit from a study examining spot size, which was suggested during technical peer review of this report. Varying the spot size may increase the likelihood of detecting a target analyte among diluents. This would require the cooperation of manufacturers to adapt sampling attachments in order to accommodate various spot sizes. During the tenure of this research, a vast increase in designer synthetic drugs has been witnessed throughout the world. As the proliferation of synthetic drugs continues, it is imperative that a method to assist law enforcement in detecting such substances is available. As these analogous substances come and go from the illicit drug market on a daily basis, the identification of them is daunting. If these drugs can, in the very least, be detected in the field using portable Raman spectroscopy, it would aid officers in their duties, even if the results could not be used in court. As the standard chemical color tests cannot differentiate between such compounds, and Raman technology can differentiate isomers, exploring the use of a portable instrumental based testing system is worth the time and consideration. A liquid sampling procedure was tested during the heroin study, however, it did not include any harsh chemicals similar to the chemical field test kits. Experimentation with graphene provided some insight into the possibility of the future creation of a non-chemical based sampling system, however, this area will need additional research. The LVMPD will continue to work towards implementing portable Raman field testing for cocaine and methamphetamine in a presumptive field testing program. The laboratory would like to continue testing methamphetamine and cocaine over the next few months, but by using an enhanced ReporteR in the lab, as well as in the field. More cocaine data is especially needed. Page 88 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 This will confirm the trend that testing accuracy increased after the device was enhanced, as is indicated by the enhanced ReporteR field data. This will also create more data to support the research findings that can be presented at an evidentiary hearing. It is also in the LVMPD’s interest to add other easily identifiable and commonly encountered drugs to the Raman-based presumptive testing program as it grows. V. References 1. Johns SH, Wist AA, Najam AR. Spot tests: A Color Chart Reference for Forensic Chemists. Journal of Forensic Sciences 1979 Jul; 24(3):631-648. 2. DEA Western Laboratory. N-Isopropylbenzylamine Hydrochloride (as “Ice” Methamphetamine Mimics) on the West Coast. Microgram Bulletin 2008 Mar. http://www.justice.gov/dea/programs/forensicsci/microgram/mg0308/mg0308.html 3. Ryder AG, O’Connor GM, Glynn TJ. Identifications and Quantitative Measurements of Narcotics in Solid Mixtures Using Near-IR Raman Spectroscopy and Multivariate Analysis. J Forensic Sci 1999; 44(5):1013-1019. 4. Eckenrode BA, Bartick, EG, Harvey SD, Vucelick ME, Wright BW, Huff RA. Portable Raman Spectroscopy for Field Analysis. Forensic Sci Comm 2001 Oct; 3(4). http://www.fbi.gov/hq/lab/fsc/current/eknrode.htm 5. Sanderson RM. Identification of N-Methylbenzylamine Hydrochloride, NEthylbenzylamine Hydrochloride, and N-Isopropylbenzylamine Hydrochloride. Microgram Journal 2008 Jan-Jun; 6. http://www.justice.gov/dea/programs/forensicsci/microgram/journal_v6_num12/pg4.html 6. SWGDRUG Recommendations. 4th Edition. Scientific Working Group for the Analysis of Seized Drugs, 2008. http://www.swgdrug.org/Documents/SWGDRUG%20Recommendations.pdf Page 89 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 7. Expert Testimony, Fed. Rules of Evid. Article 7 Rule 702, 109th Congress, 2nd Session, H.R. Prt. 9, p.13 (2006). http://www.uscourts.gov/rules/Evidence_Rules_2007.pdf 8. Witnesses, Nev. Admin. Code. NAC-50 (2003). http://leg.state.nv.us/NAC/NAC-050.html 9. Testimony by Experts, Witnesses, Nev. Rev. Stat. 4 NRS §§ 50.275 (1971). http://leg.state.nv.us/nrs/NRS-050.html#NRS050Sec275. 10. Monographs; Clarke’s Analysis of Drugs and Poisons. Pharmaceutical Press, 2004 MedicinesComplete Browser version 1.1.1586.26116 Copyright 2002. 11. http://en.wikipedia.org/wiki/File:Isopropylbenzylamine.png 12. Kovar KA, Laudszun, M. Chemistry and Reaction Mechanisms of Rapid Tests for Drugs of Abuse and Precursors Chemicals. Scientific and Technical Notes, United Nations 1989 Feb; 6 v.89-51669. 13. Harley JH, Wiberley SE. Instrumental Analysis. New York: Wiley, 1954. http://www.questia.com/read/14613301?title=Instrumental%20Analysis 14. Schrader B, Moore AH. Laser-Based Molecular Spectroscopy for Chemical Analysis – Raman Scattering Processes. Pure Appl Chem 1997; 69:1451-1468. 15. Buzzini P. Applications of Raman Spectroscopy in Forensic Sciences. Continuing Education for Forensic Professionals Program. FSI, Las Vegas, NV. June 15-20, 2009. 16. Kuptsov AH. Applications of Fourier Transform Raman Spectroscopy in Forensic Science. J Forensic Sci 1994 Mar; 39(2):305-318. 17. Bruker Optics (2010) http://www.brukeroptics.com/senterra-overview.html 18. Schulz H, Baranska M. Identification and Quantification of Valuable Plant Substances by IR and Raman Spectroscopy. Vib Spec 2007 Jan; 43(1):13-25. www.sciencedirect.com DOI:10.1016/j.vibspec.2006.06.001 Page 90 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 19. Barua AG, Hazarika S, Hussain M, Misra AK. Spectroscopic Investigation of the Cashew Nut Kernel (Anacardium Occidentale). Open Sci Food J 2008; 2:85-88. http://bentham.org/open/tofsj/openaccess2.htm DOI: 10.2174/1874256400802010085 20. Schulz H, Baranska M. Application of Raman Spectroscopy for the Analysis of Essential Oil Plants and Related Products. http://www.publish.csiro.au/?act=view_file&file_id=SA0402064.pdf 21. Schrader B, Schulz H, Andreev GN, Klump HH, Sawatzi J. Non-Destructive NIR-FTRaman Spectroscopy of Plant and Animal Tissues, of Food and Works of Art. Talanta 2000 Oct; 53(1):35-45 www.sciencedirect.com DOI: 10.1016/S0039-9140(00)00385-4 22. Schrader B, Schulz H, Baranska M, Andreev GN, Lehner C, Sawatzi J. Non-Destructive Raman Analyses – Polyacetylenes in Plants. Spectrochimica Acta Part A: Molec Biomolec Spec 2005 May; 61(7):1395-1401. www.sciencedirect.com DOI: 10.1016/j.saa.2004.10.048 23. Gierlinger N, Schwanninger M. The Potential of Raman Microscopy and Raman Imaging in Plant Research. Spectroscopy 2007; 21(2):68-89. http://iospress.metapress.com/content/x827404031624232/ 24. Schrader, B, Klump HH, Schenzel K, Schulz H. Non-destructive NIR FT Raman Analysis of Plants. J Molec Struc 1999 Oct; 509(1-3):201-212. www.sciencedirect.com DOI: 10.1016/S0022-2860(99)00221-5 25. Ryder AG. Classification of Narcotics in Solid Mixtures Using Principle Component Analysis and Raman Spectroscopy. J Forensic Sci 2002; 47(2):275-284. 26. Littleford RE, Matousek P, Towrie M, Parker AW, Dent G, Lacey RJ, Smith WE. Raman Spectroscopy of Street Samples of Cocaine Obtained Using Kerr Gated Fluorescence Rejection. Analyst 2004; 129:505-506. www.rsc.org/analyst DOI: 10.1039/b401091d 27. Radtke J, Patterson M. Validation Study for First Defender. MAFS Newsletter 2009 Spr; 38:31-47. 28. DeltaNu, Inc. (2010) http://deltanu.com/reporter/ Page 91 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 29. Hargreaves MD, Page K, Munshi T, Tomsett R, Lynch G, Edwards HGM. Analysis of Seized Drugs Using Portable Raman Spectroscopy in an Airport Environment - A Proof of Principle Study. J Raman Spec 2008; 39:873-880. www.interscience.wiley.com DOI: 10.1002/jrs.1926 30. US Patent Application No: 2006/0212275 A1 (Published Sept. 21, 2006) (Pirzer, et al, applicant). 31. Lombardi JR, Leona M, Vo-Dinh T, Antoci P. Development of Advanced Raman Spectroscopy Methods and Databases for the Evaluation of Trace Evidence and The Examination of Questioned Documents (Phase 1). US Department of Justice: Award No 2006-DN-BX-K034. 32. Nguyen D, Wu E. More Raman Analyses Less Fluorescence Interference. Spectroscopy: The Application Notebook 2011 Apr11:15. 33. Bell SEJ, Beattie JR, McGarvey JJ, Peters L, Sirimuthu NMS, Speers SJ. Development of Sampling Methods for Raman Analysis of Solid Dosage Forms of Therapeutic And Illicit Drugs. J Raman Spectrosc 2004; 35: 409-417. 34. Schulze G, et al. Investigation of Selected Baseline Removal Techniques as Candidates for Automated Implementation. Appl Spectrosc 2005; 59(5): 545-574. 35. Leger MN, Ryder AG. Comparison of Derivative Preprocessing and Automated Polynomial Baseline Correction Method for Classification and Quantification of Narcotics in Solid Mixtures. Appl Spectrosc 2006; 60(2): 182-193. 36. Wieboldt D. Understanding Raman Spectrometer Parameters. Raman Tech for Today’s Spectroscopists June 2010: 20-31. 37. Adar F. Molecular Spectroscopy Workbench: Entering Raman’s Realm. Spectroscopy March 2011; 26(3):22-25. Page 92 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 38. Appiah B, et al. A Novel Method for Practical Implementation of Shifted-Excitation Raman Difference Spectroscopy. Raman Technology for Today’s Spectroscopists June 2012: 22-33. 39. Volodin BL, et al. Quantitative and Qualitative Analysis of Fluorescent Substances and Binary Mixtures by Use of Shifted Excitation Raman Difference Spectroscopy. http://www.pd-ld.com/wp-content/uploads/2013/01/SERDS.pdf 40. Gardner C, et al. Handheld Raman Analysis in the Field for Improved Efficiency in the Lab. 2013 ASCLD Symposium Poster. Thermo Fisher Scientific. May 5-9. Durham, NC. http://www.ascld.org/wp-content/uploads/2013/06/Gardner-TruNarc-Poster.pdf 41. US Patent Application No: 8346487B2 (Published Jan. 1, 2013) (Wright, et al, applicant). 42. Color Tests. TM4-2B-4. Illinois State Police Training Manual. Illinois State Police. August 1992. 43. Markert H, et al. Evaluation of the Thermo Scientific FirstDefender RM Raman Spectrometer. US Department of Justice: Award No 2008-MU-MUK003. 44. Weston RG. Quick Screening of Crystal Methamphetamine/Methyl Sulfone Exhibits by Raman Spectroscopy. J Forensic Sci July 2010; 55 (4):1068-1075. 45. Liming X, et al. Graphene as a Substrate to Suppress Fluorescence in Resonance Raman Spectroscopy. J Amer Chem Soc July 2009; 131 (29):9890-9891. DOI: 10.1021/ja9037593. 46. Radosav SP, et al. Graphene: Substrate Preparation and Introduction. J Stuct Bio 2010 (in press). DOI:10.1013/j.jsb.2010.10.002. Reviewed References (not cited) Adar F. Molecular Spectroscopy Workbench FT-IR and Raman: A Synergism, Not Competing Technologies. Spectoscopy 2009 Oct;24(10):16-19 Brewster VL, Edwards HGM, Hargreaves MD, Munshi T. Identification of the Date Rape Drug GHB and its Precursor GBL by Raman Spectroscopy. Drug Test Anal 2009;1:25-31. www.drugtestinganalysis.com DOI: 10.1002/dta.11 Page 93 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 Washer G, Blum Jr. F, Raman Spectroscopy for the Nondestructive Testing of Carbon Fiber. Res Lett Materials Sci 2008. http://utc.mst.edu/documents/Washer_Letter_2008.pdf DOI:10.1155/2008/693207 Hodkiewicz J, Wall M. Raman Spectroscopy as a Rapid Characterization Tool for Heterogeneous Solids. Raman Technology for Today’s Spectroscopists (Spectroscopy Supplement) 2009 Jun;18-25. Moore DS, Scharff RJ. Portable Raman Explosives Detection. Anal Bioloanal Chem 2009; 393:1571-1578. http://www.springerlink.com/content/5126n134t836j336/?p=8de5977ec61349ea99d91671874 9fb01&pi=0 DOI: 10.1007/s00216-008-2499-5 Kanable R. Returning Results with FIDO Programs. Officer.com 2008 Nov. http://www.officer.com/print/Law-Enforcement-Technology/Returning-results-with-FIDOprograms/1$44228 Olds WJ, et al. Spatially Off-set Raman Spectroscopy (SORS) for the Analysis and Detection of Packaged Pharmaceuticals and Concealed Drugs. For Sci Int 2011; 9:69-77. http://www.horiba.com/fileadmin/uploads/Scientific/Documents/Raman/RA19.pdf Ken Smith. Raman Microscopy (1430). McCrone Research Institute. August 9-11, 2010. Chicago, IL. Page 94 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice. Presumptive Field Testing Using Portable Raman Spectroscopy: Research and Development on Instrumental Analysis for Forensic Science: Award Number 2010-DN-BX-K201 VI. Dissemination of Research Findings “Presumptive Raman Portable Field Testing” But Wait, There’s More! Combining Multiple Methods Into an “All-In-One” Option for Evidence Processing and Testing. National Institute of Justice Conference, June 19, 2012, Arlington, VA. The author was interviewed by Dr. Edward Bartick of Suffolk University in early 2010 in regards to preliminary laboratory testing using the ReporteR which was completed prior to the award of this grant. This information was used by Dr. Edward Bartick in a chapter titled “In Situ Crime Scene Analysis” in a book titled Infrared and Raman Spectroscopy in Forensic Science, 1st Edition, edited by Chalmers, Edwards, and Hargreaves, published by John Wiley and Sons, Ltd. in 2012. Page 95 of 95 This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice.